TW201013836A - Microprocessors, intergarated circuits and methods for reducing noises thereof - Google Patents

Microprocessors, intergarated circuits and methods for reducing noises thereof Download PDF

Info

Publication number
TW201013836A
TW201013836A TW98132275A TW98132275A TW201013836A TW 201013836 A TW201013836 A TW 201013836A TW 98132275 A TW98132275 A TW 98132275A TW 98132275 A TW98132275 A TW 98132275A TW 201013836 A TW201013836 A TW 201013836A
Authority
TW
Taiwan
Prior art keywords
substrate
voltage
clamp
wire
bias
Prior art date
Application number
TW98132275A
Other languages
Chinese (zh)
Other versions
TWI405297B (en
Inventor
Raymond A Bertram
Mark J Brazell
Vanessa S Canac
Darius D Gaskins
James R Lundberg
Matthew Russell Nixon
Original Assignee
Via Tech Inc
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
Priority claimed from US12/237,463 external-priority patent/US7920019B2/en
Priority claimed from US12/237,483 external-priority patent/US7978001B2/en
Application filed by Via Tech Inc filed Critical Via Tech Inc
Publication of TW201013836A publication Critical patent/TW201013836A/en
Application granted granted Critical
Publication of TWI405297B publication Critical patent/TWI405297B/en

Links

Abstract

A microprocessor including a substrate bias rail providing a bias voltage during a first operation mode, a supply node providing a core voltage, a clamp device coupled between the bias rail and the supply node, and control logic. The control logic turns on the clamp device to clamp the bias rail to the supply node during a second operating mode and turns off the clamp device during the first operating mode. The clamp devices may be implemented with P-channel and N-channel devices. Level shift and buffer circuits may be provided to control the clamp devices based on substrate bias voltage levels. The microprocessor may include a substrate with first and second areas each including separate substrate bias rails. The control logic separately turns on and off clamp devices to selectively clamp the substrate bias rails in the first and second areas based on various power modes.

Description

201013836 六、發明說明: 【發明所屬之技術領域】 本發明主要關於一種於微處理器晶粒(die)提供基底偏 壓(substrate biasing)以減低次臨界漏電流(sub-threshold leakage),特別係有關於一種分別箝制基底偏壓導線至核 心電壓與參考電壓以最小化裝置基底之雜訊之裝置與方 法,進而改善裝置執行性能。 【先前技術】 因互補式金氧半導體(Complementary Metal-Oxide Semiconductor,以下簡稱CMOS)電路比其他類型的積體 電路(integrated circuit,以下簡稱1C)較為密集(dense)且 其消耗的電力較少,所以CMOS技術已成為於積體電路中 之數位電路設計的主流(dominant style)。CMOS電路係由 N 通道金氧半導體(n-channel metal_oxide-semiconductor, 以下簡稱NMOS)與P通道金氧半導體p_channei metal-oxide-semiconductor,以下簡稱 PMOS)共同組成, 根據設計、比例(scale)、材質(material)及製程(process)之 不同,NMOS與PMOS分別具有一臨界電壓(此指閘極對 源極之電壓)。由於積體電路設計及製造技術不斷發展,操 作電壓及裝置尺寸也隨之降低。65微米(nanometer,nm) 製程為應用於大量CMOS半導體製程之先進光蝕刻技術 (lithographic process)且更有益於超大型積體電路(very large scale integrated circuit,以下簡稱 VLSI)之製造,如201013836 VI. Description of the Invention: [Technical Field] The present invention is mainly concerned with providing a substrate biasing on a microprocessor die to reduce sub-threshold leakage, in particular It relates to an apparatus and method for separately clamping a substrate biasing wire to a core voltage and a reference voltage to minimize noise of the device substrate, thereby improving device performance. [Prior Art] Complementary Metal-Oxide Semiconductor (CMOS) circuits are denser than other types of integrated circuits (hereinafter referred to as 1C) and consume less power. Therefore, CMOS technology has become the dominant style of digital circuit design in integrated circuits. The CMOS circuit is composed of an N-channel metal-oxide-semiconductor (hereinafter referred to as NMOS) and a P-channel metal-oxide-semiconductor (hereinafter referred to as PMOS), depending on the design, scale, and material. Between material and process, NMOS and PMOS have a threshold voltage (this refers to the voltage of the gate to the source). As integrated circuit design and manufacturing technology continues to evolve, operating voltages and device sizes are also reduced. The 65 micrometer (nm) process is an advanced lithographic process for a large number of CMOS semiconductor processes and is more beneficial for the manufacture of very large scale integrated circuits (VLSI), such as

CNTR2419I00-TW/0608-A41899-TWF 4 201013836CNTR2419I00-TW/0608-A41899-TWF 4 201013836

微處理器等。隨著裝置尺寸與電壓位準的減少,每個裝置 的通道長度與氧化層厚度(oxide thickness)也跟著減少。製 造業者已改用具有較低臨界電壓之閘極材質以增加次臨 界漏電流(sub-threshold leakage current)。當閘極對源極之 電壓低於CMOS裝置之臨界電壓時,次臨界漏電流流經汲 極(drain)與源極(source)之間。多個傳統電路之每個CMOS 的基底介面(或為井區或基底接點(bulk tie/connection)) 係耦接於對應之一電力線(例如PMOS基底接點耦接於 核心電壓VDD,NMOS基底接點耦接於參考電壓VSS)。 在此類傳統結構,次臨界漏電流在動態環境(如正常操作期 間)下可佔總耗電力的約30%或是以上之比例。 通常需要積體電路操作於低電力模式(l〇w power mode)(如睡眠模式或冬眠(hibernation)模式)與儘可能地減 少電力消耗。於低電力模式期間’偏壓產生器(bias generator)或充電幫浦(charge pump)以與供應電力不同之 電壓位準來偏壓裝置之基底。偏壓產生器可提供於晶片上 或晶片外(offchip)。另一種情況,偏壓產生器將pM〇s的 基底接點之電壓提升至高於核心電壓VDD的電壓並將 NMOS的基底接點的電壓降低至低於參考電壓να的電 壓。這樣的基底偏壓明顯減少於低電力模式下之次臨界電 壓漏電流,藉以保存電力總量。然而,在大型積體裝置(如 微處理器),需要傳送基底偏壓至分佈於晶粒上的多個裝 置。雖然有可能於晶粒上提供多個偏壓產生器,但上述多 個偏壓產生器消耗了有價值(valuable)的晶粒範圍,所以需Microprocessor, etc. As device size and voltage levels decrease, the channel length and oxide thickness of each device also decreases. Manufacturers have switched to gate materials with lower threshold voltages to increase sub-threshold leakage current. When the gate-to-source voltage is lower than the threshold voltage of the CMOS device, the sub-critical leakage current flows between the drain and the source. The base interface of each CMOS of the plurality of conventional circuits (or a well region or a bulk tie/connection) is coupled to a corresponding one of the power lines (eg, the PMOS substrate contacts are coupled to the core voltage VDD, the NMOS substrate The contact is coupled to the reference voltage VSS). In such conventional structures, the subcritical leakage current can account for about 30% or more of the total power consumption in a dynamic environment (e.g., during normal operation). Integral circuits are typically required to operate in a low power mode (such as sleep mode or hibernation mode) with as little power consumption as possible. During a low power mode, a bias generator or charge pump biases the substrate of the device at a different voltage level than the supplied power. The bias generator can be provided on-wafer or off-chip. Alternatively, the bias generator boosts the voltage of the base contact of pM〇s to a voltage higher than the core voltage VDD and lowers the voltage of the NMOS base contact to a voltage lower than the reference voltage να. Such a substrate bias is significantly reduced by the sub-critical voltage leakage current in the low power mode, thereby preserving the total amount of power. However, in large integrated devices, such as microprocessors, it is desirable to transfer the substrate bias to a plurality of devices distributed over the die. Although it is possible to provide a plurality of bias generators on the die, the above plurality of bias generators consume a valuable range of crystal grains, so

CNTR2419IOO-TW/06O8-A41899-TWF 5 201013836 要求最小化偏壓產生器的數量。基底偏壓導線儘可能距晶 粒較遠處來繞線,以傳送基底偏壓。於低電力模式,偏壓 產生器驅動基底偏壓,以最小化次臨界漏電流與降低電 力。於正常操作模式,偏壓產生器驅動偏壓導線之電壓至 對應之供應電壓,以嘗試改進裝置的執行性能。偏壓導線 分佈之相關的阻抗之位準將導致於遍佈(across)積體電路 之基底之電壓變動(voltage variation)。基底偏壓導線也會 由於電容耗合(capacitive coupling)導致引入雜訊,影響裝 置的執行性能。 在最小化電壓變動與雜訊以及維持裝置執行性能的 ,時,要求將基底偏壓導線遍布於大型積體裝置(如微處理 器)的晶粒,這是現有技術亟須解決的問題。 【發明内容】 有鑑於此,根據-實施例所述之一種微處理器 括·第-基底偏壓導線,於第一操作模式提供 偏壓。第-電源供應點提供核H至少 = 接於第一基底偏壓導線與第一供應點 置。於第二操作模式期間,控制敬置-控制裝 箝制第-基底偏壓導線至“ 位裝置導通’以 作模式期間,不導通箝位裝置源供應節點’並於第一操 置可為半導體裝置,例 一偏移電壓之第-基底偏壓。微處理器核Ϊ 路用以偏壓箝位裝置’以確保於第-操作;準移:: 一——9· 下模式期間不導適 201013836 Γ裝置1處理器可包括緩衝器用以控制多個籍位裝 導線撤二器:基包底=底偏料線舆第二基底偏*CNTR2419IOO-TW/06O8-A41899-TWF 5 201013836 It is required to minimize the number of bias generators. The substrate biasing wire is wound as far as possible from the grain to transfer the substrate bias. In low power mode, the bias generator drives the substrate bias to minimize sub-critical leakage current and reduce power. In the normal mode of operation, the bias generator drives the voltage of the biasing conductor to the corresponding supply voltage in an attempt to improve the performance of the device. The level of impedance associated with the biasing conductor distribution will result in voltage variations across the base of the integrated circuit. The substrate biased conductor also introduces noise due to capacitive coupling, affecting the performance of the device. In order to minimize voltage variations and noise and to maintain device performance, it is required to spread the substrate bias wires over the crystal grains of a large integrated device (e.g., a microprocessor), which is a problem that the prior art has to solve. SUMMARY OF THE INVENTION In view of the above, a microprocessor according to the embodiment includes a first-substrate biasing conductor that provides a bias voltage in a first mode of operation. The first power supply point provides a core H at least = connected to the first substrate bias wire and the first supply point. During the second mode of operation, the control respect-controls the clamping of the first-substrate biasing conductor to the "bit device conducting" mode, the non-conducting clamping device source supply node' and the first operation may be a semiconductor device For example, the first-substrate bias of the offset voltage. The microprocessor core is used to bias the clamp device to ensure the first operation; the quasi-shift:: one -9· during the lower mode is not suitable for 201013836 The Γ device 1 processor may include a buffer for controlling a plurality of home-mounted wire detachers: a base package bottom 底 bottom feed line 舆 a second base bias*

二基底偏μ。根據—實施例作模式期間提供第 基底偏壓相對於核心電壓具有作模式期間,第一 偏壓相對於參考電|具有—負電壓偏移偏移’而第二基底 微處理器可包括一基底 於-實施例中,在第-操作模式第=第 第一基=::r位於第—區域的半導== 第一£域之+導體裝置保持電力開啟。 二基底偏料線_位裝置。控制裝置可 = =接於f基底偏壓導線與第二基錢㈣線之箱= 根據-實施例所述之-種積體電路包括—基底、位於 基底之第-基底偏㈣線與第二基底偏導線、位於基底 之第一電源供應導體提供相對於參考電壓之核心電壓,上 述參考電壓由位於基底之第二電源供應導體所提供、位於 基底且耦接於第一電源供應導體與第一基底偏壓導線之 間之至少一第一箝位裝置、位於基底且耦接於第二電源供 應導體與第二基底偏壓導線之間之至少一第二籍位^ 置、以及一控制裝置。於積體電路之第一操作模式期間, 提供第一基底偏壓於第一基底偏壓導線,提供第二美底偏 壓於第二基底偏壓導線,其中第一基底偏壓高於=心電The second substrate is biased by μ. During the embodiment mode, during which the first substrate bias is provided with respect to the core voltage, the first bias voltage has a negative voltage offset offset with respect to the reference voltage | and the second substrate microprocessor may include a substrate In the embodiment, in the first mode of operation, the first base ==:r is located in the semi-conductance of the first region == the first £ domain + the conductor device remains powered on. Two base bias line _ bit device. The control device can == connect to the f-substrate biased wire and the second base (four) wire box = the embodiment according to the embodiment - the integrated circuit includes a substrate, a base-substrate (four) line and a second at the base The base biasing conductor, the first power supply conductor on the substrate provides a core voltage relative to a reference voltage, the reference voltage is provided by the second power supply conductor located on the substrate, located on the substrate and coupled to the first power supply conductor and the first At least one first clamping device between the substrate biasing wires, at least one second location on the substrate coupled between the second power supply conductor and the second substrate biasing wire, and a control device. Providing a first substrate biased to the first substrate biasing conductor during the first mode of operation of the integrated circuit, providing a second aesthetic bias to the second substrate biasing conductor, wherein the first substrate bias is higher than = Electricity

CNTR2419I00-TW/Q608-A41899-TWF 201013836 壓,而第二基底偏壓低於參考電壓。控制裝置具有用以控 制上述第一箝位裝置之第一輸出端,並且具有用以控制第 二箝位裝置之第二輸出端。於第一操作模式時,控制裝置 將第一與第二箝位裝置不導通,並於第二操作模式將上述 第一與第二箝位裝置導通以箝制第一基底偏壓導線至第 一電源供應導體以及箝制第二基底偏壓導線至第二電源 供應導體。 積體電路可包括位準移位電路以根據基底偏壓位準 將箝位裝置導通與不導通。積體電路可包括耦接於箝位裝 置之緩衝器。基底可分為第一與第二區域,上述區域分別 具有複數半導體裝置,其中第一基底偏壓導線與第二基底 偏壓導線與至少一第一箝位裝置位於基底之第一區域。 根據一實施例所述之一種減低微處理晶片雜訊之方 法,上述微處理晶片包括第一基底偏壓導線,用以減少次 臨界漏電流。根據一實施例,當微處理器晶片於第一電力 狀態時,第一基底偏壓導線箝制第一基底偏壓導線至核心 電壓,微處理器晶片於第二電力狀態時,不箝制第一基底 偏壓導線,並於提供第一基底偏壓至第一基底偏壓導線。 上述方法包括導通所選取的複數第一箝位裝置,上述 複數第一箝位裝置用來維持第一基底偏壓導線之電壓於 相對於上述核心電壓之變動在一第一既定最小電壓位準 且複數第一箝位裝置沿著上述第一基底偏壓導線分佈。上 述方法之步驟可包括將第一半導體裝置之汲極與源極耦 接於第一基底偏壓導線與核心電壓之間,當微處理器於第 CNTR2419IO0-TW/06O8-A41899-TWF 8 201013836 一電力狀態,導通第一半導 力狀態,不導通第一半導置:當微處理器於第二電 偏移電壓㈣動第—基底偏=上述方法可包括提供一 或是低於核心電壓,以及提導,電壓高於核心電壓 -半導體裝置之閘極至高:或第;低 上述:法可包括將第:;導體CNTR2419I00-TW/Q608-A41899-TWF 201013836 is pressed while the second substrate bias is lower than the reference voltage. The control device has a first output for controlling the first clamping device and a second output for controlling the second clamping device. In the first mode of operation, the control device disables the first and second clamping devices, and turns on the first and second clamping devices to clamp the first substrate biasing wire to the first power source in the second operating mode. The conductor is supplied and the second substrate biasing wire is clamped to the second power supply conductor. The integrated circuit can include a level shifting circuit to turn the clamping device on and off depending on the substrate bias level. The integrated circuit can include a buffer coupled to the clamping device. The substrate can be divided into first and second regions, each of the regions having a plurality of semiconductor devices, wherein the first substrate biasing wires and the second substrate biasing wires and the at least one first clamping device are located in a first region of the substrate. In accordance with an embodiment of the method of reducing micro-processed wafer noise, the micro-processed wafer includes a first substrate biasing conductor for reducing sub-critical leakage current. According to an embodiment, when the microprocessor chip is in the first power state, the first substrate bias wire clamps the first substrate bias wire to the core voltage, and the microprocessor chip does not clamp the first substrate when in the second power state The wire is biased and a first substrate biased to the first substrate biased wire is provided. The method includes conducting a plurality of selected first clamping devices, wherein the plurality of first clamping devices are configured to maintain a voltage of the first substrate biasing conductor at a first predetermined minimum voltage level with respect to the variation of the core voltage and A plurality of first clamping devices are distributed along the first substrate biasing wires. The method may include the step of coupling the drain and the source of the first semiconductor device between the first substrate bias wire and the core voltage, and when the microprocessor is in the CNTR2419IO0-TW/06O8-A41899-TWF 8 201013836 The power state, conducting the first semi-conductive state, not conducting the first semi-conducting: when the microprocessor is at the second electrical offset voltage (four), the first-substrate biasing method may include providing one or lower than the core voltage, And to mention that the voltage is higher than the core voltage - the gate of the semiconductor device is high: or the first; low above: the method may include the first: conductor

於微處理L:片線與核心電壓’以及 α 杈供級衝器,用以緩衝第一箝位致能信 k供緩衝箝位致能信號至第二半導體裝置之-閘 極。於-實施例’ _箝位致能信賴第—箝位致能信號 之電壓位準相同。 微處理器晶片可分為第一與第二區域以及可包括第 一基底偏壓導線。於一實施例,第一基底偏壓導線位於第 一區域’第二基底偏壓導線位於第二區域。在本案中,上 述方法更包括選擇箝制第一與第二基底偏壓導線至核心 電壓或者是選擇不箝制基底偏壓導線以及在微處理器的 多種電力狀態下於基底偏壓導線接收對應之偏壓。 【實施方式】 為使本發明之上述目的、特徵和優點能更明顯易懂, 下文特舉一較佳實施例,並配合所附圖式’作詳細說明如 下: 實施例: 熟悉此技藝之人士皆可由以下描述’視其實際應用與 需要’創造及使用本發明。然而,熟悉此技藝之人士皆可 9The micro-processing L: chip line and core voltage ' and α 杈 supply stage buffers are used to buffer the first clamp enable signal k for buffering the clamp enable signal to the gate of the second semiconductor device. The voltage level of the clamp-enhanced first-clamp enable signal is the same. The microprocessor die can be divided into first and second regions and can include a first substrate biasing wire. In one embodiment, the first substrate biasing wire is in the first region and the second substrate biasing wire is in the second region. In the present case, the above method further includes selecting to clamp the first and second substrate bias wires to the core voltage or selecting not to clamp the substrate bias wires and receiving corresponding offsets on the substrate bias wires in various power states of the microprocessor. Pressure. The above described objects, features and advantages of the present invention will become more apparent from the following description of the preferred embodiments. The present invention can be created and used by the following description 'depending on its actual application and needs'. However, anyone familiar with this skill can do 9

CNTR2419I00.TW/0608-A41899-TWF 201013836 變動為較佳之實施例,以應用於其他實施例。因此,本發 明的目的不只限於所顯示之實施例,也應揭露於包括與其 原則一致之廣泛範圍及新的特點。 發明人考量傳統基底偏壓於低電力模式時將裝置基 底偏壓至不同於供應電壓之電壓位準會具有明顯的阻抗 與電容雜訊耦合(capacitiven〇isec〇upling)。例如,其缺點 包括因沿著基底偏壓導線長度而增加的電壓降(v〇ltage drop)將導致基底偏壓明顯之變動,而於正常操作模式時, 耦接於裝置之雜訊使得執行性能下降。因此,發明人提供 具有基底偏壓箝制之微處理器,以減少電壓變動與雜訊耦 合’並於以下說明及結合第1圖至第8圖描述。 第1圖係顯示包括整合於p型基底1〇1上之CMOS裝 置之一積體電路100之一實施例以及根據一實施例所述之 整合於積體電路100上之基底偏壓電路1〇2之區塊圖。雖 然所顯示之特定結構為雙層井(twin well)製程,但依然可 考慮槪用其他類—型的製程(如N型井(N-well)、P型井 ((P-well)及三層井(triple well)等)。N 型井區 1〇3、105 與 107形成於P型基底101内,並且第二:N型井區105為深 N 型井區(deep N-well region)。隔離之 p 型井區(is〇iated P-well) 109形成於深N型井區105内。第一 n型井區103 用以製造P型通道裝置Π1,而隔離之P型井區109用以 製造N型通道裝置113。熟悉此技藝人士皆暸解第三n型 井區107可應用於其他裝置。雖然第1圖僅顯示二個通道 裝置111與113,熟悉此技藝人士皆瞭解任何數量之額外 10CNTR2419I00.TW/0608-A41899-TWF 201013836 Variations are preferred embodiments for use in other embodiments. Therefore, the object of the present invention is not limited to the embodiments shown, but should be disclosed to include a wide range and novel features in accordance with the principles. The inventors consider that the biasing of the substrate base to a voltage level different from the supply voltage when the conventional substrate bias is applied to the low power mode will have significant impedance and capacitive noise coupling. For example, its disadvantages include an increased voltage drop due to biasing the length of the wire along the substrate, which will cause significant variations in substrate bias, while in normal operating mode, noise coupled to the device causes performance to be performed. decline. Accordingly, the inventors have provided a microprocessor with a substrate bias clamp to reduce voltage variations and noise coupling' and are described below and in conjunction with Figures 1-8. 1 shows an embodiment of an integrated circuit 100 including a CMOS device integrated on a p-type substrate 110, and a substrate bias circuit 1 integrated in the integrated circuit 100 according to an embodiment. Block diagram of 〇2. Although the specific structure shown is a twin well process, other types of processes (such as N-well, P-well, and III) can still be considered. A well, etc.) N-type well regions 1〇3, 105, and 107 are formed in the P-type substrate 101, and the second: N-type well region 105 is a deep N-well region (deep N-well region) An isolated p-well zone 109 is formed in the deep N-well zone 105. The first n-well zone 103 is used to fabricate a P-channel device Π1, while the isolated P-well zone 109 Used to fabricate the N-channel device 113. It is understood by those skilled in the art that the third n-well region 107 can be applied to other devices. Although Figure 1 shows only two channel devices 111 and 113, those skilled in the art will be aware of any number. Extra 10

CNTR2419IO0-TW/O6O8-A41899-TWF 201013836 ' 裝置皆可應用於P型基底101上。 - 成對的 P 型擴散區(diffusion region)(P+)l 15 與 117 以 及N型擴散區(N+) 119形成P型通道裝置in於n型井區 103内。P型通道裝置U1更包括將閘極絕緣層 ; insulatorlayer)121覆蓋在P型擴散區115及117之N型井 . 區ι〇3上。p型擴散區(ρ+)ιΐ5形成為汲極端,標註為“D”; P型擴散區(Ρ+)1 Π形成為源極端,標註為“s,’;以及閘極 絕緣層121形成為閘極端,標註為“G”。根據裝置的特別 功能’ P型通道裝置1U的閘極端G與汲極端D耦接於積 體電路1〇〇的對應信號(未繪示)。p型通道裝置U1的源 極端3耦接於一核心電壓(()〇1^¥〇如辟)¥1)1)。在一實施例 中,上述核心電壓yDD由一第一電源供應節點提供。n 型擴散區119形成為一井區或基底接點作此 connection),標註為b” ’耗接於提供p型通道裝置η! 之基底偏壓VBNA之基底偏壓導線(substrate bias φ rail)104。對於Ν型通道褒置⑴,成—對的ν型擴散區 (Ν+)123及125以及ρ型擴散區(ρ+)127形成於隔離之ρ ^井區109内’而閘極絕緣層129形成在覆蓋於ν型擴散 區123及125之ρ型井區1〇9上。Ν型擴散區125形成為 〆及極端D ’ Ν型擴散區123形成為源極端S ;卩及閘極絕 緣層129形成為閘極端G。Ν型通道裝置113❸閘極端G 與/及極端根據裝置的特別功能耦接於積體電路10〇 上的對應信號(未緣示),型通道裝置113的源極端3輛 接另核〜電壓VSS,為了與上述核心電壓VDD區別’ CNTR2419I〇0-TW/__A41899twf 201013836 因此稱為參考電壓(core reference voltage)VSS ’上述參考 - 電壓VSS於實施例中為一接地信號。在一實施例中,上述 · 參考電壓VSS由一第二電源供應節點提供。P型擴散區127 形成為一井區或基底接點B,耦接於用以提供基底偏壓 VBPA於N型通道裝置113之基底偏壓導線106。 核心電壓VDD與參考電壓VSS可透過導體或是導電 : 線路等(例如熟悉此技藝人士皆瞭解之導電穿孔、導電節 點、導電導線、導電匯流排與匯流排信號等)提供於整個 積體電路或晶片。基底偏壓導線1〇4與1〇6也可透過導體 ® 或導電線路等實施。 基底偏壓電路102包括偏壓產生器112,上述偏壓產 生器112具有輸出端以分別於基底偏壓導線1〇4與1〇6上 提供基底偏壓VBNA與VBPA。雖然於實施例中偏壓產生 器II2係以位於積體電路100之電荷幫浦實施,但依然可 考慮以其他類型的電壓產生器實施。偏壓產生器112由控 制裝置114所提供之偏壓控制信號bctl控制。控制裝置 114有一輸出端,提供箝位致能信號ΕΝρ至ρ型位準移位 ❹ 電路(IMype level shifter,LSP)116之輪入端,而上述ρ型 位準移位電路116有-輪出端’提供對應的箝制移位致能 信號PEN至Ρ型通道箝位裝置pci的_。ρ型通道籍 位裝置pc!具有耦接於核心、電壓VDD麟極,其没極與 基底耦接至基底偏壓導線丨〇4。控制襞置114有另一輸出 端’提供另-n位致能信號職至㈣位準移位(N_type level shifter,LSN)電路118之輸入端,上述N型位準移位CNTR2419IO0-TW/O6O8-A41899-TWF 201013836 'The device can be applied to the P-type substrate 101. - A pair of P-type diffusion regions (P+)l 15 and 117 and an N-type diffusion region (N+) 119 form a P-type channel device in the n-type well region 103. The P-type channel device U1 further includes a gate insulating layer 121 covering the N-type wells of the P-type diffusion regions 115 and 117. The p-type diffusion region (ρ+) ιΐ5 is formed as a 汲 extreme, denoted as “D”; the P-type diffusion region (Ρ+)1 Π is formed as a source terminal, labeled “s,'; and the gate insulating layer 121 is formed as The gate terminal is marked as “G.” According to the special function of the device, the gate terminal G of the P-channel device 1U and the 汲 terminal D are coupled to corresponding signals (not shown) of the integrated circuit 1〇〇. The p-channel device The source terminal 3 of U1 is coupled to a core voltage ((), 1), 1). In an embodiment, the core voltage yDD is provided by a first power supply node. The region 119 is formed as a well region or substrate contact, and is labeled b" 'substrate bias φ rail 104 that is supplied to the substrate bias VBNA that provides the p-channel device η!. For the Ν-type channel arrangement (1), the pair-shaped ν-type diffusion regions (Ν+) 123 and 125 and the p-type diffusion region (ρ+) 127 are formed in the isolated ρ ^well region 109 and the gate insulating layer 129 It is formed on the p-type well region 1〇9 covering the v-type diffusion regions 123 and 125. The 扩散-type diffusion region 125 is formed as a 〆 and an extreme D ′ Ν type diffusion region 123 is formed as a source terminal S; and a gate insulating layer 129 is formed as a gate terminal G. The 通道-type channel device 113 is connected to the corresponding signal on the integrated circuit 10A according to the special function of the device (not shown), and the source terminal 3 of the channel device 113 is connected to the voltage VSS. In order to distinguish from the above-mentioned core voltage VDD, CNTR2419I〇0-TW/__A41899twf 201013836 is therefore referred to as a reference voltage VSS. The above-mentioned reference-voltage VSS is a ground signal in the embodiment. In an embodiment, the above reference voltage VSS is provided by a second power supply node. The P-type diffusion region 127 is formed as a well region or substrate contact B coupled to a substrate biasing conductor 106 for providing a substrate bias voltage VBPA to the N-channel device 113. The core voltage VDD and the reference voltage VSS can be transmitted through the conductor or conductive: a line or the like (for example, a conductive via, a conductive node, a conductive wire, a conductive bus bar and a bus bar signal, which are known to those skilled in the art) are provided in the entire integrated circuit or Wafer. The substrate bias wires 1〇4 and 1〇6 can also be implemented through conductors® or conductive lines. The substrate biasing circuit 102 includes a bias generator 112 having an output to provide substrate bias voltages VBNA and VBPA on the substrate biasing conductors 1〇4 and 1〇6, respectively. Although the bias generator II2 is implemented with a charge pump located in the integrated circuit 100 in the embodiment, it can be considered to be implemented with other types of voltage generators. The bias generator 112 is controlled by a bias control signal bctl provided by the control device 114. The control device 114 has an output terminal for providing a clamp enable signal ΕΝρ to a wheel-in terminal of a ρ-type level shift ❹ circuit (LSP) 116, and the ρ-type level shift circuit 116 has a round-out The terminal 'provides a corresponding clamped shift enable signal PEN to the _ type channel clamp device pci. The p-channel home device pc! has a core, a voltage VDD, and a pole is coupled to the substrate biasing wire 丨〇4. The control device 114 has another output terminal 'providing an additional -n bit enable signal to the input of the N_type level shifter (LSN) circuit 118, the N-type level shift

CNTR2419I00-IW/O608-A41899-TWF 12 201013836CNTR2419I00-IW/O608-A41899-TWF 12 201013836

- 電路U8有一輸出端,提供對應的箝制移位致能信號NEN - 至N型通道箝位裝置NC1的閘極。N型通道箝位裝置Να 的源極輕接至參考電壓VSS’其没極與基底輕接至基底偏 壓導線106。控制裝置Π4切換箝位致能信號ENp與ENN 於積體電路100之參考電壓vss與核心電壓VDD之間。 P型位準移位電路116移動箝制移位致能信號pEN之電壓 範圍於參考電壓vss與基底偏壓VBNA之間,n型位準 移位電路118移動箝制移位致能信號NEN之電壓範圍於 _ 基底偏壓VBPA與核心電壓VDD之間。通常當控制裝置 114設置(assert)箝位致能信號ENp為低位準時箝制移 位致能信號PEN設置為低位準以導通p型通道箝位裝置 PC1以箝制基底偏壓導線1〇4至核心電壓VDD。當控制裝 置114設置箝位致能信號ENP為高位準時,則二^通道 箝位裝置pci將不導通。當控制裝置114設置箝位致能信 號ENN為高位準時,則設置箝制移位致能信號nen為^ φ位準以導通N型通道箝位裝置NC1而籍制基底偏壓導線 106至參考電壓VSS。當控制裝置114設置触致能信號 ENN為低位準時,則N型通道箝位裝置NC1將不導通。 當要求積體電路100操作於低電力模式時,控制裝置 114將没置箝位致能信號enp為高位準,並設置箝位致能 i號ENN為低位準,以不導通將箝位裴置pci與不 導通。需注意的是積體電路1〇〇可能具有多個操作狀態或 操作模式,上述多個操作狀態或模式包括一或多個低電力 模式或低電力狀態。上述低電力模式是積體電路1〇〇之至- Circuit U8 has an output providing a corresponding clamped shift enable signal NEN - to the gate of N-channel clamp IC1. The source of the N-channel clamp device Να is lightly connected to the reference voltage VSS', and its pole is lightly connected to the substrate biasing wire 106. The control device Π4 switches the clamp enable signals ENp and ENN between the reference voltage vss of the integrated circuit 100 and the core voltage VDD. The P-type level shifting circuit 116 shifts the voltage range of the clamped shift enable signal pEN between the reference voltage vss and the substrate bias voltage VBNA, and the n-type level shifting circuit 118 shifts the voltage range of the clamped shift enable signal NEN. Between the substrate bias voltage VBPA and the core voltage VDD. Generally, when the control device 114 asserts the clamp enable signal ENp to a low level, the clamp shift enable signal PEN is set to a low level to turn on the p-channel clamp device PC1 to clamp the base bias wire 1〇4 to the core voltage. VDD. When the control device 114 sets the clamp enable signal ENP to a high level, the two-channel clamp device pci will not be turned on. When the control device 114 sets the clamp enable signal ENN to a high level, the clamp shift enable signal nen is set to the ^φ level to turn on the N-channel clamp device NC1 to manufacture the base bias wire 106 to the reference voltage VSS. . When the control device 114 sets the trigger enable signal ENN to a low level, the N-channel clamp device NC1 will not conduct. When the integrated circuit 100 is required to operate in the low power mode, the control device 114 sets the clamp enable signal enp to a high level, and sets the clamp enable i number ENN to a low level to disable the clamp. Pci and non-conductivity. It is noted that the integrated circuit 1 may have multiple operational states or modes of operation including one or more low power modes or low power states. The above low power mode is the integrated circuit 1

CNTR2419I00-TW/0608-A41899-TWF 13 201013836 少一部分區域操作於低電力狀態(conditi〇n)或者是關閉。 於低電力模式,控制裝置114也控制偏壓產生器丨12,並 以一第一基底偏移電壓(substrate bias offset voltage)驅動 基底偏壓VBNA以高於核心電壓VDD之電壓,並以一第 二基底偏移電壓驅動基底偏壓VBPA以低於參考電壓 VSS。根據實際的結構,第一基底偏移電壓與第二基底偏 移電壓可為等效或者是不同的電壓。亦即,於低電力模式 時,基底偏壓VBNA相對於核心電壓VDD具有一正電壓 偏移,基底偏壓VBPA相對於參考電壓vss具有一負電 ⑩ 壓偏移。因此,於低電力模式,將P型通道裝置ln之基 底電壓驅動為南於核心電壓VDD之電壓,並將N型通道 裝置113的基底電壓驅動為低於參考電壓之電壓,以 使上述二者之裝置之次臨界漏電流最小化。當需要將積體 電路100切換至正常操作模式以正常運作時,控制裝置 114將控制偏壓產生器112以驅動基底偏壓VBNA至核心 電壓VDD之電壓位準,以及驅動基底偏壓VBpA至夫考 電壓VSS之電壓位準。因此,於正常操作模式期間,;型❹ 通道裝置111之基底B驅動至核心電壓VDD,而N型通 道裝置113之基底B驅動至參考電壓vs $。 基底偏壓導線104與106繞線(r〇uted)至整合於p型基 底ιοί之每個裝置(包括N型通道裝置113與p型通道裝 置m)。基底偏壓VBNA與情A需要分別與基底偏壓 導線104及基底偏壓導線106保持一致。通常較大尺寸之 P型基底101與/或較大之積體電路(integrated devices)具有CNTR2419I00-TW/0608-A41899-TWF 13 201013836 A small portion of the area operates in a low power state (conditi〇n) or is off. In the low power mode, the control device 114 also controls the bias generator 丨12 and drives the substrate bias voltage VBNA to a voltage higher than the core voltage VDD with a first substrate bias offset voltage. The two substrate offset voltage drives the substrate bias voltage VBPA to be lower than the reference voltage VSS. Depending on the actual configuration, the first substrate offset voltage and the second substrate offset voltage may be equivalent or different voltages. That is, in the low power mode, the substrate bias voltage VBNA has a positive voltage offset with respect to the core voltage VDD, and the substrate bias voltage VBPA has a negative voltage 10 offset with respect to the reference voltage vss. Therefore, in the low power mode, the base voltage of the P-type channel device ln is driven to a voltage south of the core voltage VDD, and the base voltage of the N-channel device 113 is driven to a voltage lower than the reference voltage, so that the above two The subcritical leakage current of the device is minimized. When it is required to switch the integrated circuit 100 to the normal operation mode for normal operation, the control device 114 controls the bias generator 112 to drive the substrate bias voltage VBNA to the voltage level of the core voltage VDD, and drives the substrate bias voltage VBpA to Check the voltage level of the voltage VSS. Therefore, during the normal operation mode, the substrate B of the type channel device 111 is driven to the core voltage VDD, and the substrate B of the N-type channel device 113 is driven to the reference voltage vs. The substrate biasing wires 104 and 106 are ruffled to each device integrated into the p-type substrate ιοί (including the N-channel device 113 and the p-channel device m). The substrate bias voltages VBNA and A need to be consistent with the substrate bias wires 104 and the substrate bias wires 106, respectively. Typically a larger size P-type substrate 101 and/or larger integrated devices have

CNTR2419I00-TW/0608-A41899-TWF 14 201013836 較長的基底偏壓導線104與106。基底偏壓導線104與106 可為實體導體(physical conductor),其阻抗導致沿著遠離 偏壓產生器〗12之導線長度而漸增之電壓降。若N型通道 裝置Π3與p型通道裝置in之一者位於距離偏壓產生器 112之相對較遠者,其基底偏屢vbna與VBPA之電屢位 準將分別與核心電壓VDD與參考電壓VSS有明顳之差 異’並導致對操作機制之執行有負面的影響。再者,基底 偏屋導線104與106容易傳送由電容耦合(capacitive coupling)或類似之效應所產生之雜訊,更影響操作並降低 效能。 利用控制偏壓產生器112分別驅動基底偏壓VBNA與 VBPA之電壓位準至核心電壓VDD與參考電壓vss,並 6¾:置箝位致能信號ENP為低位準(所以箝制移位致能信號 PEN為低位準)與箝位致能信號ENN為高位準(所以箝制 移位致能信號NEN為低位準)以將積體電路1〇〇切換回正 常操作模S。以此方式’箝位裝置PC1與⑽分別籍制 基底偏壓導線104與106至核心電壓VDD與參考電壓 vss。雖然僅顯示用於基底偏壓導線1〇4之一 p型通道箝 位裝置pci以及用於基底偏壓導線1〇6之一 N型通=箝 位裝置NC1’但可使用任何數量之箝位裝置分別沿著= 導線104與106之長度而分佈。在一實施例+,箝位裝置 之數量與位置係根據箝制各基底偏壓導線相對 核心電壓VDD與參考電壓vss之既定 ’、 ♦ ! v心之既疋最小電壓位準而 定。在此方式下’當H位裝置致能時,基底偏壓導線⑽CNTR2419I00-TW/0608-A41899-TWF 14 201013836 Longer substrate bias wires 104 and 106. The substrate biasing wires 104 and 106 can be physical conductors whose impedance results in an increasing voltage drop along the length of the wire away from the bias generator 12. If one of the N-channel device Π3 and the p-channel device in is located far away from the bias generator 112, the base repeat vbna and the VBPA electrical level will be respectively related to the core voltage VDD and the reference voltage VSS. The difference between the two has led to a negative impact on the implementation of the operational mechanism. Furthermore, the substrate biasing wires 104 and 106 easily transmit noise generated by capacitive coupling or the like, which affects operation and reduces efficiency. The control bias generator 112 drives the voltage levels of the substrate bias voltages VBNA and VBPA to the core voltage VDD and the reference voltage vss, respectively, and the clamp enable signal ENP is low (so the clamp enable signal PEN is clamped) The low level) and the clamp enable signal ENN are at a high level (so the clamp shift enable signal NEN is at a low level) to switch the integrated circuit 1 回 back to the normal operation mode S. In this manner, the clamp devices PC1 and (10) respectively produce the substrate bias wires 104 and 106 to the core voltage VDD and the reference voltage vss. Although only one p-channel clamp device pci for the substrate biasing wire 1〇4 and one of the base biasing wires 1〇6 for the N-type pass=clamping device NC1' are shown, any number of clamps can be used. The devices are distributed along the length of the = wires 104 and 106, respectively. In an embodiment +, the number and position of the clamping devices are based on the minimum voltage level of the predetermined ', ♦, v center of the respective base bias voltages relative to the core voltage VDD and the reference voltage vss. In this mode, when the H-bit device is enabled, the substrate biasing wire (10)

CNTR2419I00-TW/O6O8-A41899-TWF 15 201013836 之電壓箝制為具有既定最小電麼位準之核心電麼vdd,而 基底偏壓導線106之電壓箝制為具有既定最小電壓位準之 參考電壓VSS。上述之箝制機制可減少電容輕合效應所產 生之雜訊,並最小化沿著基底偏壓導線1〇4與1〇6之電壓 變動。在一實施例,當基底偏壓導線104與1〇6箝制為核 心電壓VDD與參考電壓VSS之後,若要求雜訊更少與維 持電力,可將偏壓產生器112停止運作(Shut down)或是切 換為低電力模式。 第2圖係顯示根據一實施例所述之基底偏壓電路202 整合於具有分佈之箝位裝置的微處理器200之晶粒之區塊 圖。基底偏壓電路202大體與第1圖之基底偏壓電路1〇2 相同,類似之裝置與元件以相同標號表示。如圖所示,偏 壓產生器112具有一輸出端’分別於基底偏壓導線1〇4與 106提供基底偏壓VBNA與VBPA。基底偏壓導線1〇4與 106繞線於微處理器的晶粒,以傳送出基底偏壓VBNA與 VBPA至選取之整合於微處理器200之P型與N型通道裝 置。一實施例所示之P型通道裝置P1具有一基底接點至 基底偏壓導線104,其作法近似於第1圖之P型通道裝置 111,N型通道裝置N1具有一基底接點至基底偏壓導線 106,其作法近似於第1圖之N型通道裝置113。雖然僅 顯示一個P型通道裝置與一個N型通道裝置,但熟悉此技 藝之人士皆瞭解可於前述之近似方法,可將多個裝置提供 於微處理器200,並以基底接點耦接至適合之基底偏壓導 線104與106之一者(以圓點標示)。耦接於基底偏壓導 CNTR2419I00-TW/0608-A41899-TWF 16 201013836 線104之P型通道箝位裝置PC1、PC2 pC8沿著基底偏 壓導線104分佈’耦接於基底偏壓導線ι〇6之n型通道箝 位裝置NCI、NC2…NC8沿著基底偏壓導線1〇6分佈。各 P型通道裝置PC1-PC8之汲極與基底耦接至基底偏壓導線 104,其源極耦接至電壓VDD。各N型通道箝位裝置 NC1-NC8之汲極與基底分別耦接至基底偏壓導線1〇6,其 源極耦接至參考電壓VSS。控制裝置114提供控制信號 BCTL以控制偏壓產生器112,其操作方法近似於第j圖 應用於積體電路100的操作方法β如第2圖所示,控制裝 置114分別提供四個ρ型箝位致能信號£&gt;吓&lt;3:〇&gt;至四個ρ 型位準移位電路LSP 116之輸入端,上述ρ型位準移位電 路116輸出對應之四個箝制移位致能信號(level_shifted clamp enable signal)PEN&lt;3:0&gt;。同樣地,控制裝置 ι14 分 別提供四個N型箝位致能信號ΕΝΝ&lt;3·〇&gt;至四個N型位準 移位電路LSN 118之輸入端’上述N型位準移位電路輸出 對應之四個箝制移位致能信號NEN&lt;3:〇&gt;。 箝制移位致能信號PEN&lt;3:0&gt;分別提供至對應之p型 通道箝位裝置PC1-PC4之問極。具體的說,箝制移位致能 信號PEN&lt;3&gt;提供至P型通道箝位裝置pci之閘極;箝制 移位致能信號PEN&lt;2&gt;提供至p型通道箝位裝置pC2之閘 極;箝制移位致能信號PEN&lt;1&gt;提供至p型通道箝位裝置 PC3之閘極以及箝制移位致能信號pEN&lt;〇&gt;提供至p型通 道箝位裝置PC4之閘極。各箝制移位致能信號pEN&lt;3:〇&gt; 分別提供於對應之四個P型緩衝器2〇1之一者之一輸入The voltage of CNTR2419I00-TW/O6O8-A41899-TWF 15 201013836 is clamped to a core power having a predetermined minimum level, and the voltage of the substrate biasing conductor 106 is clamped to a reference voltage VSS having a predetermined minimum voltage level. The above-described clamping mechanism can reduce the noise generated by the capacitive coupling effect and minimize the voltage variation of the biased wires 1〇4 and 1〇6 along the substrate. In one embodiment, after the substrate bias wires 104 and 1〇6 are clamped to the core voltage VDD and the reference voltage VSS, if less noise is required and power is maintained, the bias generator 112 may be shut down or It is switched to low power mode. 2 is a block diagram showing the substrate bias circuit 202 integrated into a die of a microprocessor 200 having a distributed clamping device, in accordance with an embodiment. The substrate biasing circuit 202 is generally identical to the substrate biasing circuit 1〇2 of FIG. 1, and like devices and elements are designated by the same reference numerals. As shown, the bias generator 112 has an output terminal' that provides substrate bias voltages VBNA and VBPA at substrate biasing conductors 1 and 4, respectively. The substrate biasing leads 1 and 4 are wound around the die of the microprocessor to transfer the substrate bias voltages VBNA and VBPA to selected P-type and N-type channel devices integrated into the microprocessor 200. The P-channel device P1 shown in one embodiment has a base contact to substrate biasing wire 104, which is similar to the P-channel device 111 of FIG. 1, and the N-channel device N1 has a base contact to the substrate bias The crimping wire 106 is similar in operation to the N-channel device 113 of Fig. 1. Although only one P-channel device and one N-channel device are shown, those skilled in the art will appreciate that a plurality of devices can be provided to the microprocessor 200 and coupled to each other by a substrate contact in the foregoing approximation. One of the suitable substrate biasing wires 104 and 106 (indicated by dots). Coupling to the base biasing conductor CNTR2419I00-TW/0608-A41899-TWF 16 201013836 The P-channel clamping device PC1, PC2 pC8 of the line 104 is distributed along the substrate biasing wire 104' coupled to the substrate biasing wire ι6 The n-channel clamps NCI, NC2...NC8 are distributed along the substrate biasing wires 1〇6. The drain and the substrate of each of the P-type channel devices PC1-PC8 are coupled to the substrate biasing conductor 104, and the source thereof is coupled to the voltage VDD. Each of the N-channel clamp devices NC1-NC8 is coupled to the substrate biasing conductor 1〇6 and the source thereof to the reference voltage VSS. The control device 114 provides a control signal BCTL to control the bias generator 112, the operation method of which is similar to the operation method β applied to the integrated circuit 100 in Fig. j. As shown in Fig. 2, the control device 114 provides four pliers respectively. The bit enable signal £&gt; scares &lt;3:〇&gt; to the input terminals of the four p-type level shift circuits LSP 116, and the p-type level shift circuit 116 outputs the corresponding four clamp shift enablers Level_shifted clamp enable signal PEN&lt;3:0&gt;. Similarly, the control device ι14 provides four N-type clamp enable signals ΕΝΝ<3·〇> to the input terminals of the four N-type level shift circuits LSN 118. The N-type level shift circuit output corresponds to the output. The four clamp shift enable signals NEN &lt; 3: 〇 &gt;. The clamp shift enable signal PEN&lt;3:0&gt; is supplied to the corresponding p-type channel clamps PC1-PC4, respectively. Specifically, the clamp shift enable signal PEN&lt;3&gt; is supplied to the gate of the P-channel clamp device pci; the clamp shift enable signal PEN&lt;2&gt; is supplied to the gate of the p-channel clamp device pC2; The clamp shift enable signal PEN&lt;1&gt; is supplied to the gate of the p-channel clamp device PC3 and the clamp shift enable signal pEN&lt;〇&gt; is supplied to the gate of the p-channel clamp device PC4. Each of the clamp shift enable signals pEN&lt;3:〇&gt; is provided in one of the corresponding four P-type buffers 2〇1, respectively.

CNTR2419I00-TW/0608-A41899-TWF 17 201013836 端,P型緩衝器201並提供對應之四個緩衝箝制移位致能 * 信號BPEN&lt;3:〇&gt; 〇具體的說,缓衝箝制移位致能信號 BPEN&lt;3&gt;為箝制移位致能信號pEN&lt;3&gt;之緩衝形式 (version);緩衝箝制移位致能信號BpEN&lt;2&gt;為箝制移位致 能信號PEN&lt;2&gt;之緩衝形式;緩衝箝制移位致能信號 BPEN&lt;1&gt;為箝制移位致能信號PEN&lt;1 &gt;之緩衝形式以及緩 衝箝制移位致能信號BPEN&lt;0&gt;為箝制移位致能信號 ΡΕΝ&lt;0&gt;之緩衝形式。緩衝箝制移位致能信號bPEN&lt;3&gt;提 供至P型通道箝位裝置PC5之閘極;緩衝箝制移位致能信馨 號BPEN&lt;2&gt;提供至P型通道箝位裝置PC6之閘極;緩&amp; 箝制移位致能信號BPEN&lt;1&gt;提供至p型通道箝位裝置pc7 之閘極以及緩衝箝制移位致能信號BPEN&lt;〇&gt;提供至p型 通道箝位裝置PC8之閘極。於此方式,不論何時箝位致能 信號ENP&lt;3:0&gt;2任一者設置為低位準,其所對應之箝制 移位致能信號PEN&lt;3:0&gt;2—者將設置為低位準,並導通 對應之P型通道箝位裝置PC1-PC4,而對應之緩衝箝制移 位致能信號BPEN&lt;3:0&gt;也設置為低位準以將對應之p型通 ❹ 道箝位裝置PC5-PC8導通。例如,當箝位致能信號ENp&lt;1&gt; 設置為低位準,則箝制移位致能信號pEN&lt;1&gt;與緩衝籍制 移位致能信號BPEN&lt;l〉t設置為低位準,因此p型通道 箝位裝置PC3與PC7導通。以此方式,控制裝置114可 選擇性致能任一對P型通道箝位裝置PC1-PC8。 與前述近似之方法’箝制移位致能信號NEN&lt;:3:0&gt;分 別提供至對應之N型通道箝位裝置NC1-NC4之閘極。具CNTR2419I00-TW/0608-A41899-TWF 17 201013836 end, P-type buffer 201 and provide corresponding four buffer clamp shift enable * signal BPEN &lt; 3: 〇 > 〇 specifically, buffer clamp shift The energy signal BPEN&lt;3&gt; is a buffer version of the clamp shift enable signal pEN&lt;3&gt;; the buffer clamp shift enable signal BpEN&lt;2&gt; is a buffered form of the clamp shift enable signal PEN&lt;2&gt;; The buffer clamp shift enable signal BPEN&lt;1&gt; is a buffered form of the clamp shift enable signal PEN&lt;1&gt; and the buffer clamp shift enable signal BPEN&lt;0&gt; is a clamp shift enable signal ΡΕΝ&lt;0&gt; Buffer form. The buffer clamp shift enable signal bPEN &lt;3&gt; is supplied to the gate of the P-type channel clamp device PC5; the buffer clamp shift enable signal Xinxin BPEN&lt;2&gt; is supplied to the gate of the P-type channel clamp device PC6; The buffered shift enable signal BPEN&lt;1&gt; is supplied to the gate of the p-channel clamp device pc7 and the buffer clamp shift enable signal BPEN&lt;〇&gt; is supplied to the gate of the p-channel clamp device PC8 . In this way, whenever the clamp enable signal ENP&lt;3:0&gt;2 is set to a low level, the corresponding clamped shift enable signal PEN&lt;3:0&gt;2- will be set to a low level. And the corresponding P-channel clamp device PC1-PC4 is turned on, and the corresponding buffer clamp displacement enable signal BPEN&lt;3:0&gt; is also set to a low level to correspond to the p-type channel clamp device PC5- PC8 is turned on. For example, when the clamp enable signal ENp&lt;1&gt; is set to a low level, the clamp shift enable signal pEN&lt;1&gt; and the buffer-made shift enable signal BPEN&lt;l>t are set to a low level, and thus the p-type The channel clamp device PC3 is electrically connected to the PC 7. In this manner, control device 114 can selectively enable any pair of P-channel clamps PC1-PC8. The method of approximating the above-mentioned approximation 'the shift enable signal NEN&lt;:3:0&gt; is supplied to the gates of the corresponding N-channel clamps NC1-NC4, respectively. With

CNTR2419I00-TW/0608-A41899-TWF 18 201013836 - 體的說,箝制移位致能信號NEN&lt;3&gt;提供至N型通道箝位 - 裝置NC1之閘極;箝制移位致能信號NEN&lt;2&gt;提供至N 型通道箝位裝置NC2之閘極;箝制移位致能信號Nen&lt;1&gt; 提供至NP型通道箝位裝置NC3之閘極以及箝制移位致能 信號NEN&lt;〇&gt;提供至N型通道箝位裝置NC4之閘極。籍 制移位致能信號NE N &lt; 3 ·· 0 &gt;分別提供於對應之四個n型緩 衝器203之一者之一輸入端,N型緩衝器203提供對應之 四個緩衝箝制移位致能信號BNEN&lt;3:〇&gt;。具體的說,緩衝 • 箝制移位致能信號BNEN&lt;3&gt;為箝制移位致能信號 NEN&lt;3&gt;之緩衝形式;緩衝籍制移位致能信號bnen&lt;2&gt;為 箝制移位致能信號NEN&lt;2&gt;之緩衝形式;緩衝箝制移位致 能信號BNEN&lt;1&gt;為箝制移位致能信號NEN&lt;1&gt;之緩衝形 式以及緩衝箝制移位致能信號ΒΝΕΝ&lt;0&gt;為箝制移位致能 信號ΝΕΝ&lt;0&gt;之緩衝形式。緩衝箝制移位致能信號 BNEN&lt;3&gt;提供至N型通道箝位裝置NC5之閘極;緩衝箝 鲁 制移位致此彳§號BNEN&lt;2&gt;提供至N型通道箱位裝置NC6 之閘極;緩衝箝制移位致能信號BNEN&lt;1&gt;提供至N型通 道箝位裝置NC7之閘極以及緩衝箝制移位致能信號 ΒΝΕΝ&lt;0&gt;提供至N型通道箝位裝置NC8之閘極。以此方 式,不論何時將箝位致能信號ENN&lt;3:〇&gt;2任一者設置為 咼位準,其所對應之箝制移位致能信號之一者 將《χ置為同位準,以將其所對應之N型通道箝位裝置 NC1-NC4 $通,而對應之緩衝箝制移位致能信號 BNEN&lt;3.0&gt;之一者也设置$高位準,以將對應之n型通道CNTR2419I00-TW/0608-A41899-TWF 18 201013836 - In other words, the clamp shift enable signal NEN&lt;3&gt; is provided to the N-channel clamp-device NC1 gate; the clamp shift enable signal NEN&lt;2&gt; Provided to the gate of the N-channel clamp device NC2; the clamp shift enable signal Nen&lt;1&gt; is provided to the gate of the NP-type channel clamp device NC3 and the clamp shift enable signal NEN&lt;〇&gt; is provided to N The gate of the type channel clamp NC4. The system shift enable signal NE N &lt; 3 ·· 0 &gt; is respectively provided at one of the input terminals of one of the four n-type buffers 203, and the N-type buffer 203 provides four buffer clamp shifts corresponding thereto. The bit enable signal BNEN &lt; 3: 〇 &gt;. Specifically, the buffering/displacement shift enable signal BNEN&lt;3&gt; is a buffering form of the clamp shift enable signal NEN&lt;3&gt;; the buffer shift enable signal bnen&lt;2&gt; is a clamp shift enable signal The buffering form of NEN&lt;2&gt;; the buffer clamp shift enable signal BNEN&lt;1&gt; is a buffering form of the clamp shift enable signal NEN&lt;1&gt; and the buffer clamp shift enable signal ΒΝΕΝ&lt;0&gt; is a clamp shift Can signal ΝΕΝ &lt;0&gt; buffer form. The buffer clamp shift enable signal BNEN &lt;3&gt; is provided to the gate of the N-channel clamp device NC5; the buffer clamp is braked to cause the 彳§ BNEN&lt;2&gt; to provide the gate to the N-type channel tank device NC6 The buffer clamp shift enable signal BNEN &lt;1&gt; is provided to the gate of the N-channel clamp device NC7 and the buffer clamp shift enable signal ΒΝΕΝ&lt;0&gt; is supplied to the gate of the N-channel clamp device NC8. In this way, whenever one of the clamp enable signals ENN&lt;3:〇&gt;2 is set to the 咼 level, one of the corresponding clamped shift enable signals will be set to the same level. The corresponding N-channel clamp device NC1-NC4 $ is connected, and the corresponding buffer clamp shift enable signal BNEN &lt;3.0&gt; is also set to a high level to correspond to the n-type channel.

CNTR2419I0Q-TW/0608-A41899-TWF 19 201013836 拼位裝置NC5-NC8之一者導通。例如,當控制裝置 設置箝位致能信號ENN&lt;24高位準,則箝制移位致能信 ' 號NEN&lt;2&gt;與緩衝箝制移位致能信號bnEN&lt;2&gt;也設置為 ' 高位準’以將N型通道箝位裝置NC2與NC6導通。以此 方式,控制裝置114可選擇性致能任一對N型通道箝位装 置 NC1-NC8。 雖然第2圖只顯示八個P型通道箝位裝置pci_pc8輿 八個N型通道箝位裝置NC1_NC8。但熟悉此技藝之人士 可根據實際積體電路100之尺寸與架構來使用任何數量的 . 通道箝位裝置與對應之箝位致能信號。同時,所顯示有關 P型通道裝置P1的信號與p型通道箝位裝置以及有關坟 型通道裝置N1的信號與N型通道箝位裝置之群級 (grouping)可為任意的,雖然僅顯示上述裝置,熟悉此技藝 之人士亦可考量多個可能的變動。例如,由控制裝置114 提供單一箝位控制信號,於移動位準之後可根據箝位襞 置的數量要求,以提供所要求的緩衝次數。同時,雖然第 2圖顯示箝位裝置PC1_PC4為共同群組,但是上述裝置可 馨 分別位於實際要求之位置(如相近於對應之裝置)£&gt;例如, ,立裝置PCi與PC2雖然彼此互相相近,但是實際上卻 疋分離(separated)的,同時於微處理器2〇〇之晶粒上,箝 位裝置PC1與PC5可實際鄰近(cl〇sed)。利用多個箝位控 制信號於微處理器200之部分選擇區域,可以選擇性致能 箝制之操作。於一實施例,沿著基底偏壓導線1〇4與1〇6 之箝位裝置的數量與實際位置由動態模擬或類似之方式CNTR2419I0Q-TW/0608-A41899-TWF 19 201013836 One of the placement devices NC5-NC8 is turned on. For example, when the control device sets the clamp enable signal ENN &lt; 24 high level, the clamp shift enable signal 'NEN&lt;2&gt; and the buffer clamp enable signal bnEN&lt;2&gt; are also set to 'high level' The N-channel clamps NC2 and NC6 are turned on. In this manner, control device 114 can selectively enable any pair of N-channel clamp devices NC1-NC8. Although Figure 2 shows only eight P-channel clamps pci_pc8舆 eight N-channel clamps NC1_NC8. However, those skilled in the art can use any number of channel clamps and corresponding clamp enable signals depending on the size and architecture of the actual integrated circuit 100. Meanwhile, the grouping of the signal indicating the P-type channel device P1 and the p-channel clamping device and the signal about the grave-channel device N1 and the N-channel clamping device can be arbitrary, although only the above is displayed. Devices, those skilled in the art, may also consider a number of possible variations. For example, a single clamp control signal is provided by control device 114, which may be based on the number of clamps required to provide the desired number of buffers after moving the level. Meanwhile, although FIG. 2 shows that the clamp devices PC1_PC4 are a common group, the above devices may be respectively located at actual requirements (eg, similar to the corresponding devices). For example, the vertical devices PCi and PC2 are similar to each other. However, in reality, it is separated, and on the die of the microprocessor 2, the clamping devices PC1 and PC5 can be physically adjacent (cl〇sed). The clamping operation can be selectively enabled using a plurality of clamp control signals in selected portions of the microprocessor 200. In one embodiment, the number and actual position of the clamping devices that bias the wires 1〇4 and 1〇6 along the substrate are dynamically simulated or similar.

^NTR2419I00-TW/0608-A41899-TWF 20 201013836 - 決定以維持雜訊位準於一最小位準,藉以取得微處理器 . 200之最佳化執行性能。 如前述之積體電路1〇〇之近似方法,微處理器2〇〇有 多個操作狀態或操作模式。上述多個操作狀態或模式包括 一或多個低電力模式或低電力狀態,而上述低電力模式係 指選擇性使微處理器2〇〇之至少一部分於低電力狀態或是 不工作°多個箝位裝置,包括箝位裝置PC1-PC8與 NC1_NC8 ’上述箝位裝置沿著基底偏壓導線104與106分 佈及橫跨遍佈於微處理器200之基底。於微處理器2〇〇的 正常操作模式期間,控制裝置114將導通或致能全部之箝 位裝置,或是被選擇之箝位裝置,以分別箝制基底偏壓導 線104與106至核心電壓VDD與參考電壓vss。於正常 操作模式,控制裴置U4關閉將偏壓產生器112關閉或者 是設定偏壓產生器112為低電力狀態,或者是控制偏壓產 生器112以分別驅動基底偏壓VBNA與VBpA至核心電壓 φ VDD與參考電壓VSS之電壓位準。控制裝置114先將所 有箝位裝置不導通或者是選擇其中之至少一者為不導 通,則可設置微處理器於低電力模式或低電力狀態。接下 來,控制裝置114致能或者是控制偏壓產生器U2以一第 一基底偏移電壓驅動基底偏壓VBNA至高於核心電壓 VDD之電壓,以及以一第二基底偏移電壓驅動基底偏壓 VBPA至低於參考電壓vss之電壓。第一與第二基底偏移 電壓可為相同或不同的電壓位準。為將微處理器由低電力 模式拉回正常操作模式,控制裝置114需先控制偏壓產生^NTR2419I00-TW/0608-A41899-TWF 20 201013836 - Decided to maintain the noise level at a minimum level to achieve optimal performance of the microprocessor. The microprocessor 2 has a plurality of operational states or operational modes as described above for the approximate method of the integrated circuit. The plurality of operational states or modes include one or more low power modes or low power states, and the low power mode refers to selectively causing at least a portion of the microprocessor 2 to be in a low power state or not to operate. Clamping devices, including clamping devices PC1-PC8 and NC1_NC8', are distributed along the substrate biasing wires 104 and 106 and across the substrate of the microprocessor 200. During the normal mode of operation of the microprocessor 2, the control device 114 will turn on or enable all of the clamping devices, or the selected clamping device, to clamp the substrate bias wires 104 and 106 to the core voltage VDD, respectively. With reference voltage vss. In the normal mode of operation, the control device U4 turns off the bias generator 112 or sets the bias generator 112 to a low power state, or controls the bias generator 112 to drive the substrate bias voltages VBNA and VBpA to the core voltage, respectively. The voltage level of φ VDD and the reference voltage VSS. The control device 114 may set the microprocessor to a low power mode or a low power state by first turning off all of the clamp devices or selecting at least one of them to be non-conductive. Next, the control device 114 enables or controls the bias generator U2 to drive the substrate bias voltage VBNA to a voltage higher than the core voltage VDD with a first substrate offset voltage, and to drive the substrate bias voltage with a second substrate offset voltage. VBPA to a voltage lower than the reference voltage vss. The first and second substrate offset voltages can be the same or different voltage levels. In order to pull the microprocessor back from the low power mode to the normal operating mode, the control device 114 first needs to control the bias generation.

CNTR2419I00-TW/0608-A41899-TWF 21 201013836 器112 ’以分別將基底偏壓導線1〇4與i〇6之基底偏壓 · VBNA與VBPA驅動回核心電壓vdd與參考電壓VSS。 . 接下來’控制裝置114導通所有箝位裝置導通或至少一箝 位裝置。如之前所述’控制裝置n4設置所有箝位致能信 號£灿&lt;3:0&gt;與ENN&lt;3:〇&gt;,或者是選擇箝位致能信號 £阶&lt;3:0&gt;與ENN&lt;3:0&gt;之至少一者來設置,以導通或是不 導通箝位裝置PC1-PC8與NC1_NC8之至少一者。 第3圖係顯示根據本發明一實施例所述之一 p型位準 移位電路LSP 116。P型位準移位電路LSP 116包括反相 粵 器301、四個p型通道裝置、p2、p3與p4、以及N型 通道裝置N1、N2、N3與N4°P型通道裝置PI、P2、P3 與P4分別具有耦接至用以提供基底偏壓VBNA之基底偏 壓導線104之源極與内部(internal)基底,n型通道裝置 Nl、N2、N3與N4分別具有耦接至參考電壓VSS之源極 與内部基底。箝位致能信號ENP可提供給P型通道裝置 P1的閘極與反相器301的輸入端。p型通道裝置pi的汲 極耦接N型通道裝置N1的汲極與閘極與N型通道裝置 ® N2的閘極。反相器301的輸出端耦接p型通道裝置p2的 閘極,上述P型通道裝置P2的汲極耦接N型通道裝置N2 的没極以及P型通道裝置P3與N型通道裝置N3的閘極。 P型通道裝置P3的汲極輕接N型通道褒置N3的沒極以及 P型通道裝置P4與N型通道裝置N4的閘極。p型通道裝 置P4與N型通道裝置N4的汲極耦接在一起,並輸出箝 制移位致能信號PEN。在操作時,輸入之箝位致能信號CNTR2419I00-TW/0608-A41899-TWF 21 201013836 112' to bias the substrate biasing wires 1〇4 and i〇6, respectively, VBNA and VBPA back to core voltage vdd and reference voltage VSS. Next, the control device 114 turns on all of the clamping devices to conduct or at least one of the clamping devices. As previously described, 'control device n4 sets all clamp enable signals £3 &gt;3:0&gt; and ENN&lt;3:〇&gt;, or selects the clamp enable signal for the order &lt;3:0&gt; and ENN&lt;; 3: 0&gt; at least one of the settings to turn on or not turn on at least one of the clamp devices PC1-PC8 and NC1_NC8. Figure 3 is a diagram showing a p-type level shifting circuit LSP 116 according to an embodiment of the invention. The P-type level shifting circuit LSP 116 includes an inverted 301, four p-channel devices, p2, p3 and p4, and N-channel devices N1, N2, N3 and N4°P-type channel devices PI, P2. P3 and P4 respectively have a source and an internal substrate coupled to a substrate biasing conductor 104 for providing a substrate bias voltage VBNA, and the n-channel devices N1, N2, N3 and N4 are respectively coupled to a reference voltage VSS. The source and internal substrate. The clamp enable signal ENP is supplied to the gate of the P-channel device P1 and the input of the inverter 301. The drain of the p-channel device pi is coupled to the drain and gate of the N-channel device N1 and the gate of the N-channel device ® N2. The output end of the inverter 301 is coupled to the gate of the p-type channel device p2, and the drain of the P-type channel device P2 is coupled to the pole of the N-channel device N2 and the P-channel device P3 and the N-channel device N3. Gate. The drain of the P-type channel device P3 is lightly connected to the N-type channel and the gate of the N3 and the gate of the P-channel device P4 and the N-channel device N4. The p-type channel device P4 is coupled to the drain of the N-type channel device N4 and outputs a clamp shift enable signal PEN. Input clamp enable signal during operation

CNTR2419I00-TW/0608-A41899-TWF 22 201013836 ' ENP將設置於參考電壓VSS與核心電壓vdd之間。而輸 • 出之箝制移位致能信號PEN之信號將設置於參考電壓 VSS與基底偏壓VBNA之間。當箝位致能信號ENP信號 設置為參考電壓VSS,P型通道裝置卩丨導通且p型通道 裝置P2不導通(反相器3〇1的輸出為核心電壓VDD)。P 型通道裝置P1推動N型通道裝置N2的閘極之位準上升 至基底偏壓VBNA’因此N型通道裝! N2將導通。N型 通道裝置N2推動P型通道裝置P3&amp;N型通道裝置N3的 籲 閘極至參考電壓VSS,因此將導通P型通道裝置P3而不 導通N型通道裝置N3°P型通道裝置P3推動p型通道裝 置P4與N型通道裝置N4的閘極至基底偏壓VBNA,將 導通N型通道裝置N4與不導通P型通道裝置p4。因此, 當箝位致能信號ENP設置為參考電壓vsS ’透過N型通 道裝置N4將使箝制移位致能信號pen之信號為參考電壓 VSS。當箱位致能信號ENP設置為核心電壓vdD,P型通 0 道裝置P1不導通而P型通道裝置P2導通。由於p型通道 裝置P1為不導通,N型通道裝置N1將推動N型通道裝 置N2的閘極為低位準’所以N型通道装置N2將不導通。 P型通道裝置P2推動P型通道裝置P3與n型通道裝置 N3的閘極至基底偏麼VBNA ’則P型通道裝置P3不導通 而N型通道裝置N3導通。N型通道裝置N3推動P型通 道裝置P4與N型通道裝置N4的閘極至參考電壓VSS, 將導通P型通道裝置P4而不導通N型通道裝置N4。因 此’當箝位致能信號ENP信號設置為核心電壓VDD,pCNTR2419I00-TW/0608-A41899-TWF 22 201013836 ' ENP will be set between reference voltage VSS and core voltage vdd. The signal of the clamped shift enable signal PEN outputted between the reference voltage VSS and the substrate bias voltage VBNA. When the clamp enable signal ENP signal is set to the reference voltage VSS, the P-channel device is turned on and the p-channel device P2 is turned off (the output of the inverter 3〇1 is the core voltage VDD). The P-type channel device P1 pushes the level of the gate of the N-type channel device N2 up to the substrate bias voltage VBNA' so the N-channel is mounted! N2 will be turned on. The N-type channel device N2 pushes the gate of the P-type channel device P3&amp;N-type channel device N3 to the reference voltage VSS, so that the P-type channel device P3 will be turned on without turning on the N-channel device N3°P-type channel device P3 to push p The gate-to-substrate bias voltage VBNA of the type channel device P4 and the N-type channel device N4 will turn on the N-channel device N4 and the non-conducting P-channel device p4. Therefore, when the clamp enable signal ENP is set to the reference voltage vsS', the signal of the clamp shift enable signal pen is made to pass through the N-type channel device N4 as the reference voltage VSS. When the tank enable signal ENP is set to the core voltage vdD, the P-type pass device P1 is not turned on and the P-type channel device P2 is turned on. Since the p-type channel device P1 is non-conducting, the N-type channel device N1 will push the gate of the N-type channel device N2 to a very low level so that the N-channel device N2 will not conduct. The P-type channel device P2 pushes the gate of the P-channel device P3 and the n-channel device N3 to the substrate VBNA', then the P-channel device P3 is non-conducting and the N-channel device N3 is turned on. The N-type channel device N3 pushes the gates of the P-type channel device P4 and the N-type channel device N4 to the reference voltage VSS, and turns on the P-type channel device P4 without conducting the N-type channel device N4. Therefore, when the clamp enable signal ENP signal is set to the core voltage VDD, p

CNTR2419I00-TW/0608-A41899-TWF 23 201013836 型通道裝置P4推動箝制移位致能信號pen之信號至基底 偏壓VBNA。在這種方式下,箝位致能信號ENP切換於參 考電壓VSS與核心電壓VDD之間,則輸出箝制移位致能 信號PEN切換於參考電壓VSS與基底偏壓VBNA之間。CNTR2419I00-TW/0608-A41899-TWF 23 The 201013836 type channel device P4 pushes the signal of the clamped enable signal pen to the substrate bias voltage VBNA. In this manner, the clamp enable signal ENP is switched between the reference voltage VSS and the core voltage VDD, and the output clamp shift enable signal PEN is switched between the reference voltage VSS and the substrate bias voltage VBNA.

第4圖係顯示根據本發明之一實施例所述之一 N型位 準移位電路LSN 118。N型位準移位電路LSN Π8包括一 反相器401,四個p型通道裝置P1、p2、p3與P4以及四 個N型通道裝置Nl、N2、N3與N4。P型通道裝置P1、 P2、P3與P4分別具有耦接至核心電壓VDD之源極與内 部基底。N型通道裝置Nl、N2、N3與N4分別具有耦接 至提供基底偏壓VBPA之基底偏壓導線1〇6之源極與内部 基底。箝位致能信號ENN可提供給N型通道裝置N1的 閘極與反相器401的輸入端β型通道裝置pi的汲極與閘 極耦接N型通道裝置N1的汲極與p型通道裝置P2的閘 極。反相器401的輸出端耦接至n型通道裝置N2的閘極, 上述N型通道裝置N2的汲極耦接至P型通道裝置P2的 汲極與P型通道裝置P3與N型通道裝置N3的閘極。P 型通道裝置P3的汲極耦接至N型通道裝置]Sf3的汲極以 及P型通道裝置P4與N型通道裝置N4的閑極。p型通道 裝置P4與N型通道裝置N4的汲極耦接在一起,並且輸 出箝制移位致能信號NEN信號。在操作中,輸入之箝位 致能信號ENN信號設置為參考電壓vss與核心電壓VDD 之間。而輸出之箝制移位致能信號NEN之信號設置於基 底偏壓VBPA與核心電壓VDD之間。當箝位致能信號ENN CNTR2419I00-TW/0608-A41899-TWF 24 201013836Figure 4 is a diagram showing an N-type alignment shift circuit LSN 118 according to an embodiment of the present invention. The N-type level shifting circuit LSN Π8 includes an inverter 401, four p-type channel devices P1, p2, p3 and P4, and four N-type channel devices N1, N2, N3 and N4. P-channel devices P1, P2, P3, and P4 have a source and an inner substrate coupled to a core voltage VDD, respectively. The N-type channel devices N1, N2, N3, and N4 respectively have a source and an internal substrate coupled to a substrate biasing conductor 1〇6 that provides a substrate bias voltage VBPA. The clamp enable signal ENN can be supplied to the gate of the N-type channel device N1 and the input terminal of the inverter 401. The drain and gate of the β-channel device pi are coupled to the drain and p-channel of the N-channel device N1. The gate of device P2. The output end of the inverter 401 is coupled to the gate of the n-type channel device N2, and the drain of the N-type channel device N2 is coupled to the drain of the P-channel device P2 and the P-channel device P3 and the N-channel device. The gate of N3. The drain of the P-type channel device P3 is coupled to the drain of the N-channel device]Sf3 and the idler of the P-channel device P4 and the N-channel device N4. The p-type channel device P4 is coupled to the drain of the N-type channel device N4, and the output clamps the shift enable signal NEN signal. In operation, the input clamp enable signal ENN signal is set between the reference voltage vss and the core voltage VDD. The signal of the output clamp shift enable signal NEN is set between the base bias voltage VBPA and the core voltage VDD. When the clamp enable signal ENN CNTR2419I00-TW/0608-A41899-TWF 24 201013836

- 設置為核心電壓VDD,N型通道裝置N1導通且N型通道 . 裝置N2不導通(反相器401的輸出為參考電壓VSS)。N 型通道裝置N1推動P型通道裝置P2的閘極至基底偏壓 VBPA,因此P型通道裝置P2導通。P型通道裝置P2推 動P型通道裝置P3及N型通道裝置N3的閘極至核心電 壓VDD,因此P型通道裝置P3不導通而N型通道裝置 N3導通。N型通道裝置N3推動P型通道裝置P4與N型 通道裝置N4的閘極至基底偏壓VBPA,因此N型通道裝 • 置N4不導通且P型通道裝置P4導通。因此,當箝位致能 信號ENP信號設置為核心電壓VDD,透過P型通道裝置 P4推動的箝制移位致能信號NEN之信號為核心電壓 VDD。當箝位致能信號ENN設置為參考電壓VSS,將不 導通N型通道裝置N1而導通N型通道裝置N2。由於N 型通道裝置N1為不導通,P型通道裝置P1推動P型通道 裝置P2的閘極為高位準,所以P型通道裝置P2不導通。 N型通道裝置N2推動P型通道裝置P3與N型通道裝置 ® N3的閘極至基底偏壓VBPA,將導通P型通道裝置P3而 不導通N型通道裝置N3。P型通道裝置P3推動P型通道 裝置P4與N型通道裝置N4的閘極至核心電壓VDD,將 不導通P型通道裝置P4而導通N型通道裝置N4。因此, 當箝位致能信號ENN設置為參考電壓VSS,N型通道裝 置N4推動箝制移位致能信號NEN信號為基底偏壓 VBPA。在這種方式下,箝位致能信號ENN切換於參考電 壓VSS與核心電壓VDD之間,且箝制移位致能信號NEN CNTR2419I00-TW/0608-A41899-TWF 25 201013836 切換於基底偏壓VBPA與核心電壓VDD之間。 請參考回第1圖’當偏壓產生器112驅動基底偏壓 VBNA為高於核心電壓vdD之電壓,P型位準移位電路 116將確保P型通道箝位裝置PC1於低電力模式下完全不 導通。更具體的說’當偏壓產生器1!2驅動基底偏壓VBNA 高於核心電壓VDD時,控制裝置114將設置箝位致能信 號ENP之位準至核心電壓VDD,並使P型通道箝位裝置 PC1不導通。若箝位致能信號ENP直接提供給P型通道箝 位裝置PC1之閘極,則上述p型通道箝位裝置PC1之閘 極電位將僅位於核心電壓VDD而其汲極之電位將高於核 心電壓VDD,可能使得p型通道箝位裝置pci部分導通。 但是’經P型位準移位電路116驅動箝制移位致能信號 PEN至基底偏壓VBNA的電壓位準,所以P型通道箝位 裝置PC1的閘極與汲極都位於高於核心電壓VDD之基底 偏壓VBNA的電壓位準,確保p型通道箝位裝置pci完 全不導通。當偏壓產生器112驅動基底偏壓VBPA為低於 參考電壓VSS之電壓,N型位準移位電路118將確保N 型通道箝位裝置NC1於低電力模式下,完全不導通。更具 體的說’當偏壓產生器112驅動基底偏壓VBNA低於參考 電壓VSS時,控制裝置114將設置箝位致能信號ENN之 位準至參考電壓VSS以不導通N型通道箝位裝置NC1。 若箝位致能信號ENN直接提供給N型通道箝位裝置NC1 之閘極’上述N型通道箝位裝置NC1之閘極之電位將僅 位於參考電壓VSS且其汲極之電位將低於參考電壓- Set to core voltage VDD, N-type channel device N1 is turned on and N-type channel. Device N2 is not turned on (the output of inverter 401 is reference voltage VSS). The N-type channel device N1 pushes the gate of the P-type channel device P2 to the substrate bias voltage VBPA, so that the P-type channel device P2 is turned on. The P-type channel device P2 pushes the gate of the P-channel device P3 and the N-channel device N3 to the core voltage VDD, so that the P-channel device P3 is not turned on and the N-channel device N3 is turned on. The N-channel device N3 pushes the gate of the P-channel device P4 and the N-channel device N4 to the substrate bias voltage VBPA, so that the N-channel device N4 is non-conducting and the P-channel device P4 is turned on. Therefore, when the clamp enable signal ENP signal is set to the core voltage VDD, the signal of the clamp shift enable signal NEN pushed through the P-channel device P4 is the core voltage VDD. When the clamp enable signal ENN is set to the reference voltage VSS, the N-channel device N1 will not be turned on and the N-channel device N2 will be turned on. Since the N-type channel device N1 is non-conducting, the P-type channel device P1 pushes the gate of the P-type channel device P2 to a very high level, so the P-type channel device P2 is not turned on. The N-type channel device N2 pushes the gate of the P-channel device P3 and the N-channel device ® N3 to the substrate bias voltage VBPA, which turns on the P-channel device P3 and does not conduct the N-channel device N3. The P-type channel device P3 pushes the gate of the P-channel device P4 and the N-channel device N4 to the core voltage VDD, and will not turn on the P-channel device P4 to turn on the N-channel device N4. Therefore, when the clamp enable signal ENN is set to the reference voltage VSS, the N-type channel device N4 pushes the clamp shift enable signal NEN signal to the substrate bias voltage VBPA. In this manner, the clamp enable signal ENN is switched between the reference voltage VSS and the core voltage VDD, and the clamp shift enable signal NEN CNTR2419I00-TW/0608-A41899-TWF 25 201013836 is switched to the substrate bias voltage VBPA and Between the core voltage VDD. Referring back to FIG. 1 'When the bias generator 112 drives the substrate bias voltage VBNA to be higher than the core voltage vdD, the P-type level shifting circuit 116 will ensure that the P-channel clamp device PC1 is completely in the low power mode. Not conductive. More specifically, when the bias generator 1!2 drives the substrate bias voltage VBNA higher than the core voltage VDD, the control device 114 sets the level of the clamp enable signal ENP to the core voltage VDD, and clamps the P-channel. The bit device PC1 is not turned on. If the clamp enable signal ENP is directly supplied to the gate of the P-channel clamp device PC1, the gate potential of the above-mentioned p-channel clamp device PC1 will be only at the core voltage VDD and the potential of the drain will be higher than the core. The voltage VDD may cause the p-channel clamp device to be partially turned on. However, 'the P-type level shifting circuit 116 drives the voltage level of the clamp shift enable signal PEN to the substrate bias voltage VBNA, so the gate and the drain of the P-channel clamp device PC1 are both higher than the core voltage VDD. The voltage level of the substrate bias voltage VBNA ensures that the p-channel clamp device pci is completely non-conductive. When the bias generator 112 drives the substrate bias voltage VBPA to a voltage lower than the reference voltage VSS, the N-type level shifting circuit 118 will ensure that the N-channel clamp unit NC1 is not conducting at all in the low power mode. More specifically, when the bias generator 112 drives the substrate bias voltage VBNA lower than the reference voltage VSS, the control device 114 sets the level of the clamp enable signal ENN to the reference voltage VSS to turn off the N-channel clamp device. NC1. If the clamp enable signal ENN is directly supplied to the gate of the N-channel clamp device NC1, the potential of the gate of the N-channel clamp device NC1 will be only at the reference voltage VSS and the potential of the drain will be lower than the reference. Voltage

CNTR2419IOO-TW/0608-A41899-TWF 26 201013836 ~ VSS,可能使得N型通道箝位裝置NCI部分導通。但是, . 經N型位準移位電路Π8驅動箝制移位致能信號NEN至 基底偏壓VBPA之電壓位準,所以N型位準移位電路Π8 的閘極與汲極之電位都低於參考電壓VSS之基底偏壓 VBPA的電壓位準,確保N型通道箝位裝置NC1不導通。 接下來’參考第2圖,當基底偏壓導線104之基底偏 壓VBNA驅動至高於核心電壓VDD之電壓位準,而對應 之至少一箝位致能信號ENP&lt;3:0&gt;設置為高位準,P型位準 肇 移動電路116分別移動對應之箝制移位致能信號 PEN&lt;3:0&gt;以確保一或多個P型通道箝位裝置PC1_PC4完 全不導通。P型緩衝器電路201驅動緩衝箝制移位致能信 號BPEN&lt;3:0&gt;至參考電壓VSS與基底偏壓VBNA之間之 位準移位電壓區,以確保當緩衝箝制移位致能信號 BPEN&lt;3:0&gt;設置為高位準時’箝位裝置;PC5-PC8也完全不 導通。同樣的,當基底偏壓導線106之基底偏壓VBPA驅 ❹動至低於參考電壓VSS之電壓位準,而對應之至少一箝位 致能信號ENN&lt;3:〇&gt;s置為低位準,N型位準移動電路us 分別移動對應之箝制移位致能信號NEN&lt;3:0&gt;以確保一或 多個N型通道箝位裝置NC1-NC4完全不導通。]Si型緩衝 器203驅動緩衝箝制移位致能信號3&gt;^&gt;^&lt;3:0&gt;至核心電 壓VDD與基底偏壓VBPA之間之位準移位電壓區,以確 保當緩衝箝制移位致能信號BNEN&lt;3:0&gt;設置為低位準 時,箝位裝置NC5-NC8也完全不導通。 第5圖係顯示根據本發明之一實施例所述之一 p型緩CNTR2419IOO-TW/0608-A41899-TWF 26 201013836 ~ VSS, may cause the NCI channel clamp NCI part to be turned on. However, the N-type level shifting circuit Π8 drives the clamped shift enable signal NEN to the voltage level of the substrate bias voltage VBPA, so the potentials of the gate and the drain of the N-type level shifting circuit Π8 are lower than The voltage level of the substrate bias voltage VBPA of the reference voltage VSS ensures that the N-channel clamp device NC1 is not conducting. Next, referring to FIG. 2, when the substrate bias voltage VBNA of the substrate bias wire 104 is driven to a voltage level higher than the core voltage VDD, the corresponding at least one clamp enable signal ENP&lt;3:0&gt; is set to a high level. The P-type register shift circuit 116 respectively moves the corresponding clamp shift enable signal PEN&lt;3:0&gt; to ensure that one or more of the P-channel clamps PC1_PC4 are completely non-conducting. The P-type buffer circuit 201 drives the buffer clamp shift enable signal BPEN&lt;3:0&gt; to a level shift voltage region between the reference voltage VSS and the substrate bias voltage VBNA to ensure that the buffer clamp shift enable signal BPEN&lt;; 3: 0&gt; set to high-level on-time 'clamping device; PC5-PC8 is also completely non-conducting. Similarly, when the substrate bias voltage VBPA of the substrate bias wire 106 is driven to a voltage level lower than the reference voltage VSS, the corresponding at least one clamp enable signal ENN&lt;3: 〇&gt;s is set to a low level. The N-type level shifting circuit us respectively moves the corresponding clamped shift enable signal NEN&lt;3:0&gt; to ensure that one or more of the N-channel clamps NC1-NC4 are completely non-conducting. The Si-type buffer 203 drives the buffer clamp shift enable signal 3&gt;^&gt;^&lt;3:0&gt; to a level shift voltage region between the core voltage VDD and the substrate bias voltage VBPA to ensure buffer clamping When the shift enable signal BNEN &lt;3:0&gt; is set to the low level, the clamp devices NC5-NC8 are also completely non-conductive. Figure 5 is a diagram showing a p-type retardation according to an embodiment of the present invention.

CNTR2419I00-TW/0608-A41899-TWF 27 201013836 衝器2:1。箝制移位致能㈣pEN信號提供至^通道裝 置Ρ1 ” Ν型通道裂置N1的問極。通道裝置Η的源 2α基Ϊ耦接至基底偏壓導線1〇4(提供基底偏壓 )’型通道裝置Ρ1的汲極耦接至Ν型通道裝置N1 的沒極π、㈣道裝置P1與N型通钱置N1躲極耦接 至P型通道裝置ΡβΝ型通道裝置N2的閘極。p型通道 裝置Ρ2的源極與基底_接至基底偏壓導線I.?型通道 裝置Ρ2的及極輕接至Ν型通道裝置Ν2的汲極。Ν型通 道裝置Ν1 ^Ν2的祕祕至參考電壓VSS’P型通道裝 置P2與N朗道裝置Μ的飾形祕浦制移位致能 信號BPEN。在操作制下,當驅動箝制移诚能信號PEN 之信號為參考電壓vss時,p型通道裝置ρι與N型通道 裝置N2都將導通,同時p型通道裝置p2與N型通道裝 置N1不導通,所以緩衝箝制移位致能信號BPEN將驅動 至參考電壓VSS。當箝制移位致能信號PEN信號為基底 偏壓VBNA時’ p型通道裝置pi與n型通道裝置N2都 不導通’同時P型通道裝置P2與N型通道裝置N1都為 導通’以推動緩衝箝制移位致能信號BPEN至基底偏壓 VBNA。在此方式下,緩衝箝制移位致能信號βρεν與箝 制移位致能信號PEN具有相同邏輯狀態,並切換於參考電 壓VSS與基底偏壓VBNA之位準移位電壓區之間。 第6圖係顯示根據本發明之一實施例所述之一 N型緩 衝器203。箝制移位致能信號NEN之信號提供給P型通道 裝置P1與N型通道裝置N1的閘極。P型通道裝置P1的 CNTR2419IO0-TW/0608-A41899-TWF 28 201013836 源極麵接至核心電壓VDD與p型通道裝置P1祕極耦接CNTR2419I00-TW/0608-A41899-TWF 27 201013836 Punch 2:1. Clamping Shift Enable (4) pEN signal is supplied to the channel device Ρ1" Ν-type channel cleave N1. The source device 2α Ϊ of the channel device Ϊ is coupled to the substrate biasing wire 1〇4 (providing the substrate bias) The drain of the channel device Ρ1 is coupled to the gate Ν of the 通道-type channel device N1, the (four) channel device P1 and the N-type money-passing device N1 are coupled to the gate of the P-channel device ΡβΝ-type channel device N2. The source and the substrate of the channel device Ρ2 are connected to the base biasing wire I. The channel device Ρ2 and the pole is connected to the 汲-type channel device Ν2. The 通道-channel device Ν1 ^Ν2 secret to the reference voltage VSS'P type channel device P2 and N Landau device Μ 秘 秘 秘 移位 移位 移位 移位 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 VSS VSS VSS VSS VSS VSS VSS VSS VSS VSS VSS Both the device ρι and the N-type channel device N2 will be turned on, while the p-type channel device p2 and the N-type channel device N1 are not turned on, so the buffer clamp shift enable signal BPEN will be driven to the reference voltage VSS. When the shift enable signal is clamped When the PEN signal is the substrate bias voltage VBNA, neither the p-channel device pi nor the n-channel device N2 At the same time, both the P-channel device P2 and the N-channel device N1 are turned on to push the buffer clamp shift enable signal BPEN to the substrate bias voltage VBNA. In this manner, the buffer clamp shift enable signal βρεν and the clamp shift The bit enable signal PEN has the same logic state and is switched between the reference voltage VSS and the level shift voltage region of the substrate bias voltage VBNA. Fig. 6 shows an N-type buffer according to an embodiment of the present invention. The signal of the clamp shift enable signal NEN is supplied to the gates of the P-channel device P1 and the N-channel device N1. The CNTR2419IO0-TW/0608-A41899-TWF 28 of the P-channel device P1 28138138 The core voltage VDD is coupled to the p-type channel device P1

^型通道裝置N1的料。N型通道裝置m的源極與 基絲接於基底偏壓導線⑽(提供給基底偏壓vbpa)。p 型通道裝置P1與N型通道裝置m喊極減至p型通 道裝置P2與N型通道|置N2的閘極。p型通道裝置 的源極祕至核心電壓VDD與P型通道裝置P2的没極耦 接至N型通道裝置N2的没極βΝ型通道裝置n2的源極 與基底耦接至基底偏壓導線1G6以及p型通道裝置p2的 沒極與N型通道裝置N2的汲極形成緩衝箝制移位致能信 號BNEN信號。在操作機财,當推㈣鄉位致能信號 NEN之信號至基底偏壓卿㈣^型通道裝置^❹ 型通道裝置N2都將導通’同時p型通道裝置p2與N型 通道裝置N1不導通,所㈣動緩衝箝制移位致能信號 BNEN至基底偏壓VBPA。當推動箝制移位致能信號NEN 至核心電壓VDD時’ P型通道裝置ρι與N型通道裝置 N2都不導通’同_ p型通道裝置p2冑N型通道裝置川 都為導通,以推動緩衝箝制移位致能信號BNEN至核心電 壓VDD。在此方式下’緩衝箝制移位致能信號bNEN與 箱制移位致能信號NEN具有相同邏輯狀態以及緩衝箝制 移位致能信號BNEN切換於核心電壓VDD與基底偏壓 VBPA之位準移位電壓區之間。 第7圖係顯示根據一實施例所述之整合於微處理器 700之晶粒之選擇區域之基底偏壓電路706,上述微處理 器包括分佈的多個箝位裝置。在一實施例中,把微處理器The material of the type channel device N1. The source and base of the N-channel device m are connected to the substrate biasing wire (10) (provided to the substrate bias vbpa). The p-type channel device P1 and the N-type channel device m are shredded to the gate of the p-type channel device P2 and the N-channel |N2. The source of the p-type channel device is connected to the source biasing of the core voltage VDD and the P-type channel device P2 to the source and the substrate of the P-type channel device n2 of the N-channel device N2 to the substrate biasing wire 1G6 And the drain of the p-type channel device p2 and the drain of the N-channel device N2 form a buffer clamp shift enable signal BNEN signal. In the operation of the machine, when the (four) township enable signal NEN signal to the base bias (4) ^ channel device ^ ❹ type channel device N2 will be turned on 'the p-channel device p2 and N-channel device N1 non-conducting (4) The dynamic buffer clamps the shift enable signal BNEN to the substrate bias voltage VBPA. When the clamp shift enable signal NEN is pushed to the core voltage VDD, the 'P-channel device ρ and the N-channel device N2 are not turned on'. The _p-channel device p2胄N-channel device is turned on to push the buffer. The shift enable signal BNEN is clamped to the core voltage VDD. In this manner, the 'buffer clamp shift enable signal bNEN has the same logic state as the tank shift enable signal NEN and the buffer clamp shift enable signal BNEN is switched to the level shift of the core voltage VDD and the substrate bias VBPA. Between voltage zones. Figure 7 is a diagram showing a substrate biasing circuit 706 integrated into a selected region of a die of a microprocessor 700, including a plurality of distributed clamping devices, in accordance with an embodiment. In an embodiment, the microprocessor

CNTR2419IO0^TW/0608-A41899-TWF 29 201013836 700为成四個區域或是象限(quadrants)7〇i,702,703與 - 704。於此實施例,於微處理器7〇〇之象限7〇4之偏壓裝 置為基底偏壓電路706。於實施例所示,於低電力模式時, 基底偏壓電路706用以偏壓位於微處理器7〇〇之象限7〇4 之裝置。基底偏壓電路706近似於第2圖之基底偏壓電路 202’以及完全位於或大體上位於微處理器7〇〇之象限7〇4 中。基底偏壓電路706包括用以偏壓位於象限7〇4之多個 P型通道裝置726之第一基底偏壓導線7〇8以及用以偏壓 位於象限704之多個N型通道裝置728之第二基底偏壓導 φ 線710。上述通道裝置726與728之架構係相似於第i圖 之P型通道裝置111 型通道裴置113。上述多個卩型 通道裝置726與N型通道裝置728分別具有多個基底接點 耦接於基底偏壓導線708與710。用簡單的形式(如方塊) 顯示裝置726與728與其基底接點至基底偏壓導線7〇8與 71〇。雖然熟悉此技藝之人士皆知上述多個p型通道裝/置 726與N型通道裝置728分佈遍及象限7〇4之區域,但是 於圖中依然顯示於象限704之一邊緣。 _ 於此所顯示之實施例,其他裝置705(如多個p型通道 裝置與N型通道裝置)分佈於微處理器7〇〇之象限 701-703。於一低電力模式時,當象限7〇4之襞置726與 728停止運作,其他裝置705仍電力開啟與被致能 (active)。象限7〇4之外部(outside)之任一或多個其他裝置 7〇5,可根據微處理器700之實際架構,具有分離的基底 偏壓電路或者是不具有分離的基底偏壓電路。於一實施例CNTR2419IO0^TW/0608-A41899-TWF 29 201013836 700 is in four regions or quadrants 7〇i, 702, 703 and - 704. In this embodiment, the biasing means in the quadrant 7〇4 of the microprocessor 7 is the substrate biasing circuit 706. As shown in the embodiment, in the low power mode, the substrate bias circuit 706 is used to bias the device located in the quadrant 7〇4 of the microprocessor 7. The substrate biasing circuit 706 is similar to the substrate biasing circuit 202' of Figure 2 and is located entirely or substantially in the quadrant 7〇4 of the microprocessor 7. The substrate biasing circuit 706 includes a first substrate biasing conductor 7〇8 for biasing a plurality of P-type channeling devices 726 in quadrants 7〇4 and a plurality of N-type channeling devices 728 for biasing the quadrants 704. The second substrate biases the φ line 710. The architecture of the above-described channel devices 726 and 728 is similar to the P-channel device 111 type channel arrangement 113 of Figure i. The plurality of 卩-type channel devices 726 and N-type channel devices 728 respectively have a plurality of substrate contacts coupled to the substrate bias wires 708 and 710. In a simple form (e.g., a square), display devices 726 and 728 are bonded to their substrate to substrate biasing conductors 7A and 71A. Although it is known to those skilled in the art that the plurality of p-type channel mounts/sets 726 and N-channel means 728 are distributed throughout the region of quadrants 7〇4, they are still shown at one edge of quadrant 704. _ In the embodiment shown herein, other devices 705 (e.g., a plurality of p-channel devices and N-channel devices) are distributed in quadrants 701-703 of microprocessor 7. In a low power mode, when the quadrants 726 and 728 of the quadrant 7 〇 4 cease to operate, the other devices 705 are still powered on and enabled. Any one or more other devices 7〇5 of the outside of the quadrant 7〇4 may have a separate substrate bias circuit or a separate substrate bias circuit depending on the actual architecture of the microprocessor 700. . In an embodiment

CNTR2419I00-TW/0608-A41899-TWF 30 201013836 τ止運作模式時,將分職供分離的基底偏壓電 .=4他象限秦彻,以偏壓上述象限之基底。於另一 實施例中,微處理器700之其他裝置7〇5之任一者可形 ^為必要電路㈣ieal path)之—部分,並且無須提供基 /屋電路給這㈣置或者使得基底偏壓電路 (disable) 〇 夕個P型通道箝位裝置712耦接於基底偏壓導線708 • / f心電壓VDD之間,多個_通道箝位裝置714耗接 於基底偏壓導線710與參考電壓vss之間。於一實施例 i通道箝位裝置712之架構與操作方法分別等同於 第2圖中微處理器200之P型通道箝位裝置pci_pc8,n 型通道箝位裝置714之架構與操作方法分別等同於微處理 器200之N型通道箝位裝置⑽-⑽,其中用簡單的形 式(如圓圈符號)顯示箝位裝置712與714。微處理器7〇〇 ^括中央控制裝置7G7,上述中央控制裝置透過對應控制 ❷仏號CCTL控制象限控制(QC)裝置716。雖然所顯示的中 央控制裝置707位於象限702,但是於微處理器7〇〇之任 何位置皆可放置中央控制裝置7〇7。象限控制裝置爪提 供控制信號QCTL,以控制偏驗生器_718,上述驗 產生器操作方法近似於前述的偏壓產生器112,並具有輸 出端’分別於基底偏壓導線7〇8與71〇形成基底偏壓VBpA 與VBNA。象限控制裝置716提供箝位致能信號ENN與 ENP至位準移位電路72〇。位準移位電路72〇包括p型位 準移位電路(未繪示)與N型位準移位電路(未繪示),上述p CNTR2419I00-TW/0608-A41899-1 31 201013836 型與N型位準移位電路分別近似於前述的位準移位電路 116 ^ 118 ’用以分別轉換由象限控制裝置716輸出之箝 位致此信號ENN與ENP為箝制移位致能信號NEN與 PEN。於所顯示的實施例’箝制移位致能信號NEN最後控 $所有P型通道籍位裝置712,而箝制移位致能信號PEN 最,控制所有n型通道箝位m14t)P型緩衝h(pb)722 /口著箝制移位致能信號PEN之信號線分佈,以滿足多個位 置所要求緩衝藉制移位致能信號PEN。同樣地,N型緩衝 器(NB)724沿著箝制移位致能信號nen之信號線分佈,以❹ 滿足多個位置所要求的緩衝箝制移位致能信 號 NEN 〇 基底偏壓電路706操作方法近似於前述的基底偏壓電 路202。於正常操作模式,當於象限704之裝置726與728 電力開啟(power up) ’象限控制裝f 716係指示偏壓產生 器718驅動基底偏壓導線7〇8與分別至核心電壓vDd 與參考電壓vss之電壓位準。象限控制裝置716設置籍位 致能信號ENN與ENP,以㈣位裝置712與714導通, 並分別箝制基底偏壓導線7〇8與71〇至核心電壓VDD與⑩ 參考電壓VSS。根據上述’位準移位電路72〇設置箝制移 位致能NEN與PEN至位準移位(shift)之電壓位準。 若有需求’則基底偏壓產生器718可為不導通或位於低電 力模式。於低電力模式,當於象限7〇4之裝置726與728 電力關閉(power down) ’象限控制裝置7丨6設置箝位'致能 信號ENN與ENP ’以將箝位裝置712與714不導通以及 位準移位電路720設置箝制移位致能信號NEN與pEN信CNTR2419I00-TW/0608-A41899-TWF 30 201013836 When the operation mode is set to τ, the base bias voltage for the separation is divided. = 4 He quadrant, to bias the base of the above quadrant. In another embodiment, any of the other devices 7〇5 of the microprocessor 700 can be formed as part of the necessary circuit (ie), and there is no need to provide a base/house circuit for the (four) or biasing the substrate. Disconnecting a P-channel clamp device 712 coupled between the substrate bias wires 708 • / f core voltage VDD, the plurality of channel clamp devices 714 are consuming the substrate bias wires 710 and the reference Between voltage vss. The architecture and operation method of the channel clamp device 712 in an embodiment are respectively equivalent to the P-channel clamp device pci_pc8 of the microprocessor 200 in FIG. 2, and the structure and operation method of the n-channel clamp device 714 are respectively equivalent to N-channel clamps (10)-(10) of microprocessor 200, wherein clamps 712 and 714 are shown in a simple form, such as a circle symbol. The microprocessor 7 includes a central control unit 7G7 which controls the quadrant control (QC) unit 716 via a corresponding control nickname CCTL. Although the central control unit 707 is shown in quadrant 702, central control unit 7〇7 can be placed at any location in microprocessor 7. The quadrant control device jaws provide a control signal QCTL to control the bias detector _718, which operates in a manner similar to the aforementioned bias generator 112 and has an output terminal 'biasing the conductors 7〇8 and 71 respectively on the substrate 〇 Forms the substrate bias voltages VBpA and VBNA. Quadrant control device 716 provides clamp enable signals ENN and ENP to level shift circuit 72A. The level shifting circuit 72A includes a p-type level shifting circuit (not shown) and an N-type level shifting circuit (not shown), and the above-mentioned p CNTR2419I00-TW/0608-A41899-1 31 201013836 type and N The type level shifting circuits are respectively approximated by the aforementioned level shifting circuits 116^118' for respectively converting the clamps outputted by the quadrant control means 716 to cause the signals ENN and ENP to be the clamped shift enable signals NEN and PEN. In the illustrated embodiment, the clamp shift enable signal NEN is finally controlled by all P-channel home devices 712, while the clamp enable signal PEN is clamped to control all n-channel clamps m14t) P-type buffer h ( Pb) 722 / port clamps the signal line distribution of the shift enable signal PEN to meet the buffered shift enable signal PEN required for a plurality of locations. Similarly, the N-type buffer (NB) 724 is distributed along the signal line of the clamp shift enable signal nen to satisfy the buffer clamp shift enable signal NEN required by the plurality of positions. The method approximates the aforementioned substrate bias circuit 202. In normal mode of operation, devices 726 and 728 in quadrant 704 power up 'quadrant control device f 716 to indicate that bias generator 718 drives substrate bias wires 7 〇 8 and to core voltage vDd and reference voltage, respectively. The voltage level of vss. The quadrant control device 716 sets the home enable signals ENN and ENP to conduct the (four) bit devices 712 and 714, and clamps the substrate bias wires 7〇8 and 71〇 to the core voltages VDD and 10, respectively, to the reference voltage VSS. The voltage level at which the shift enable NEN and PEN are shifted to the level shift is set in accordance with the above-described level shift circuit 72. The substrate bias generator 718 can be non-conducting or in a low power mode if desired. In the low power mode, when the devices 726 and 728 in quadrant 7 电力 power down the 'quadrant control device 7 丨 6 sets the clamp 'enable signals ENN and ENP ' to disable the clamping devices 712 and 714 And the level shift circuit 720 sets the clamp shift enable signal NEN and the pEN letter

CNTR2419I00-TW/0608-A41899-TWF 32 201013836 ‘ 號至位準移位之電壓位準。於前述之近似方式,象限控制 . 裝置716係指示偏壓產生器718驅動基底偏壓導線7〇8至 高於核心電壓VDD之一基底偏壓以及驅動基底偏壓導線 710至低於參考電壓VSS之一基底偏壓。因此,於低電力 模式,可減少次臨界漏電流並將符位裝置722與724完全 關閉。以此方式,當微處理器700之象限704有效的停止 運作,於象限701-703的部份裝置或全部裝置仍保持電力 開啟或致能。 _ 熟悉此技藝之人士皆知可能會有多個的變動。中央控 制裝置707可位於微處理器700的任何位置,且可控制其 他基底偏壓電路(未繪示),上述其他基底偏壓電路近似於 基底偏壓電路706且位於微處理器700上。例如,其他象 限701-703分別可包括一近似之基底偏壓電路,並利用中 央控制裝置707控制上述基底偏壓電路,用以偏壓一或多 個其他裝置705。雖然顯示的基底偏壓電路706用以偏壓 φ 位於微處理器700之實際象限區704之裝置,但基底偏壓 電路706可調整偏壓的對應範圍及位置二者之一,以偏壓 微處理器700的任何選擇的範圍或者是區域(如1/8、1/4、 1/2及3/4等)之裝置。同時,任何數量之基底偏壓電路皆 可用以偏壓位於微處理器7〇〇之選擇區域之裝置。在一實 施例中’多個基底偏壓電路可共用一個偏壓產生器。 第8圖係顯示根據一實施例所述之分為多個區域之微 處理器之區塊圖’上述區域分別包括分佈的箝位裝置與基 底偏壓電路。中央控制裝置802提供控制信號CTL1、CTL2 33CNTR2419I00-TW/0608-A41899-TWF 32 201013836 The voltage level of the ‘number to the level shift. In the foregoing approximate manner, quadrant control. Device 716 instructs bias generator 718 to drive substrate biasing conductor 7〇8 to a substrate bias higher than core voltage VDD and drive substrate biasing conductor 710 to below reference voltage VSS. A substrate bias. Therefore, in the low power mode, the subcritical leakage current can be reduced and the bit devices 722 and 724 are fully turned off. In this manner, when quadrant 704 of microprocessor 700 effectively ceases to function, some or all of the devices in quadrants 701-703 remain powered on or enabled. _ People familiar with this art know that there may be multiple changes. The central control unit 707 can be located anywhere in the microprocessor 700 and can control other substrate bias circuits (not shown) that are similar to the substrate bias circuit 706 and located at the microprocessor 700. on. For example, the other quadrants 701-703 can each include an approximate substrate biasing circuit and control the substrate biasing circuit with a central control device 707 for biasing one or more other devices 705. Although the substrate bias circuit 706 is shown to bias φ the device located in the actual quadrant region 704 of the microprocessor 700, the substrate bias circuit 706 can adjust one of the corresponding ranges and positions of the bias voltage to bias Any selected range of microprocessors 700 is either a device of an area (e.g., 1/8, 1/4, 1/2, and 3/4, etc.). At the same time, any number of substrate biasing circuits can be used to bias the devices located in selected regions of the microprocessor 7A. In one embodiment, a plurality of substrate biasing circuits may share a bias generator. Figure 8 is a block diagram showing a microprocessor divided into regions according to an embodiment. The above regions respectively include distributed clamp means and a base bias circuit. Central control unit 802 provides control signals CTL1, CTL2 33

CNTR2419IOO-TW/O608-A41899-TWF 201013836 與CTL3以控制基底偏壓,分別用於區域804、806與808 之裝置。控制信號CTL1控制區域804之基底偏壓電路 810 ;控制信號CTL2控制區域806之基底偏壓電路816 以及控制信號CTL3控制區域808之基底偏壓電路822。 基底偏壓電路810、816與822分別近似於第7圖之基底 偏壓電路706,用以提供基底偏壓至對應之每個區域之成 對的基底偏壓導線。以此方式,基底偏壓電路810提供基CNTR2419IOO-TW/O608-A41899-TWF 201013836 and CTL3 to control substrate bias, respectively for devices 804, 806 and 808. The base bias circuit 810 of the control signal CTL1 control region 804; the base bias circuit 816 of the control signal CTL2 control region 806 and the base bias circuit 822 of the control signal CTL3 control region 808. Substrate biasing circuits 810, 816 and 822, respectively, approximate substrate biasing circuit 706 of Figure 7 for providing a substrate biased to a corresponding pair of substrate biasing conductors for each of the regions. In this manner, the substrate bias circuit 810 provides a base.

底偏壓,用於區域804之P型通道裝置812與N型通道裝 置814 ;基底偏壓電路816提供基底偏壓,用於區域806 之P型通道裝置818與N型通道裝置820以及基底偏壓電 路822提供基底偏壓,用於區域808之P型通道裝置824 與N型通道裝置826。P型通道箝位裝置與N型通道箝位 裝置分別用以麵接於每個區域804,806與808之基底偏 壓導線,以及分別利用基底偏壓電路81〇,gw與822控 制上述箝位裝置之操作方法(未顯示於第8圖中),近似於 則述之偏壓電路706之操作方法。以此方式,中央控制裝 置802忐選擇性停止運作於任一或多個區域與 808之裝置’其中於被停止運作之區域中,對應基底偏壓 電路提供基錢壓至對㈣置,崎止運作上述區域,並 且最小化次臨界漏電流。同時,當區域m、8()6與議 之任者彳τ止運作’具有位準移位電路之基底偏壓電路將 使箱位裝置完全不導通。然而,當區域8〇4、腸與應 =-者致能’將導通對應箝位裝置,以分別箝制基底偏 壓導線至核4壓VDD與參考電壓他以最小化雜訊。Bottom bias, P-channel device 812 and N-channel device 814 for region 804; substrate bias circuit 816 provides substrate bias for P-channel device 818 and N-channel device 820 and substrate for region 806 Bias circuit 822 provides a substrate bias for P-channel device 824 and N-channel device 826 for region 808. The P-channel clamp device and the N-channel clamp device are respectively used to face the base bias wires of each of the regions 804, 806 and 808, and the clamps are controlled by the base bias circuits 81, gw and 822, respectively. The method of operation of the bit device (not shown in Figure 8) approximates the method of operation of the bias circuit 706 described above. In this manner, the central control unit 802 忐 selectively stops the operation of the device in any one or more of the regions 808 'where it is in the region where the operation is stopped, and the corresponding base bias circuit provides the base voltage to the pair (four), The above areas are operated and the sub-critical leakage current is minimized. At the same time, when the region m, 8 () 6 and the speaker 彳 τ are stopped, the substrate bias circuit having the level shift circuit will make the tank device completely non-conductive. However, when the region 8〇4, the intestine and the should be enabled, the corresponding clamping device will be turned on to clamp the substrate biasing wire to the core 4 to VDD and the reference voltage, respectively, to minimize the noise.

CNTR2419I00-TW/0608-A41899-TWF 34 201013836 - 前述之任一實施例皆可應用於更多類型之架構,參考 • 電壓(如VSS)可近似於〇伏特(Volts,V)與核心電壓(如 VDD)可近似於1V。在一實施例中,偏壓產生器驅動一 8〇〇 毫伏(mil1 volts,mV)之偏移電壓(offset voltage)分別至對 應之核心電壓位準以及參考電壓位準。於一實施例中,於 低電力模式期間,當核心電壓VDD為lv,基底偏壓VBNA 則近似於1.8V以及當參考電壓VSS為0V,基底偏壓VBPA 則近似於_800毫伏。根據裝置的操作模式,可變化實際的 核心電壓。例如’於實際架構模式或實際狀態之下,核心 電壓VDD可變動在近似於5〇〇mV至14V之間。在一實 施例中’基底偏壓VBNA之偏移電壓可不同於基底偏壓 VBPA之偏移電壓’例如’偏移電壓分別為3〇〇mV與5〇〇 mV。於任何事件中,偏壓產生器112分別驅動基底偏壓 VBNA與VBPA之基底偏壓導線1〇4與1〇6 一偏移電壓其 相對於對應的核心電壓與參考電壓。 _ 於一實施例之正常操作模式期間,箝位裝置沿著基底 偏壓導線放置以確保當箝位裝置致能時,每一基底偏壓導 線之電壓由核心電壓以及參考電壓變動的範圍不會超過 一既定最小電壓位準。於一實施例中,該既定最小電壓位 準近似於10mV。於一實施例中,由核心電壓以及參考電 壓變動的既定最小電壓位準不相同。根據實際應用裝置(如 積體電路100或者是微處理器2〇〇、7〇〇與8〇〇)的架構與 ^數以決定該既定最小電壓位準。可使用任何方法(如數學 換型分析或動態模擬等)決定箝位裝置之位置,以確保基底CNTR2419I00-TW/0608-A41899-TWF 34 201013836 - Any of the foregoing embodiments can be applied to more types of architectures, reference • Voltage (such as VSS) can be approximated to Volts (V) and core voltage (eg VDD) can be approximated to 1V. In one embodiment, the bias generator drives an offset voltage of 8 volts volts (mV) to a corresponding core voltage level and a reference voltage level, respectively. In one embodiment, during the low power mode, when the core voltage VDD is lv, the substrate bias voltage VBNA is approximately 1.8V, and when the reference voltage VSS is 0V, the substrate bias voltage VBPA is approximately _800 millivolts. The actual core voltage can be varied depending on the mode of operation of the device. For example, the core voltage VDD can vary between approximately 5 〇〇 mV and 14 V under actual architectural mode or actual state. In one embodiment, the offset voltage of the substrate bias VBNA may be different from the offset voltage of the substrate bias voltage VBPA, e.g., the offset voltage is 3 〇〇 mV and 5 〇〇 mV, respectively. In any event, bias generator 112 drives substrate bias voltage VBNA and VBPA substrate bias conductors 1〇4 and 1〇6, respectively, with an offset voltage relative to the corresponding core voltage and reference voltage. During the normal mode of operation of an embodiment, the clamping device is placed along the substrate bias wire to ensure that the voltage of each substrate biasing wire varies from core voltage to reference voltage when the clamping device is enabled. More than a predetermined minimum voltage level. In one embodiment, the predetermined minimum voltage level is approximately 10 mV. In one embodiment, the predetermined minimum voltage levels that are varied by the core voltage and the reference voltage are not the same. The predetermined minimum voltage level is determined according to the architecture and number of actual application devices (e.g., integrated circuit 100 or microprocessors 2, 7 and 8). The position of the clamp can be determined by any method (such as mathematical analysis or dynamic simulation) to ensure the base

CNTR2419I00-TW/O608-A41899-TWF 35 201013836 偏壓導線之偏壓分別相對於核心電壓vdd盥參考電壓 VSS的變動維持在該既定最小電壓位準的範圍内。 在其他實施例,基底偏壓可由晶片外提供,所以積體 電路或晶片基底可以包括偏壓產生器或者是不包括偏壓 產生器。例如’積體電路1〇〇或微處理器2〇〇可不包括偏 壓產生器112’因此基底偏麗VBNA與VBPA由外部提 供。同樣地,微處理器700不包括偏壓產生器718,以及 微處理器800之基底偏壓電路810、816與822之任一咬 多個電路。當微處理器700未包括偏壓產生器時,由於抑 制裝置依然控制箝位裝置以及對應之電路,因此會有大體 相同之動作。 本發明雖以較佳實施例揭露如上,然其並非用以限 定本發明的區域,任何熟習此項技藝者,在不脫離本發明 之精神和區内,當可做些許的更動與潤飾,因此本發明之 保護區當視後附之申請專利範圍所界定者為準。CNTR2419I00-TW/O608-A41899-TWF 35 201013836 The bias voltage of the biasing conductor is maintained within the range of the predetermined minimum voltage level with respect to the fluctuation of the core voltage vdd盥 reference voltage VSS, respectively. In other embodiments, the substrate bias may be provided off-wafer, so the integrated circuit or wafer substrate may include a bias generator or may not include a bias generator. For example, the integrated circuit 1 or the microprocessor 2 may not include the bias generator 112' so that the substrate biases VBNA and VBPA are externally supplied. Similarly, microprocessor 700 does not include bias generator 718, and any of substrate bias circuits 810, 816, and 822 of microprocessor 800 bite multiple circuits. When the microprocessor 700 does not include a bias generator, there is substantially the same behavior since the suppressor still controls the clamp and the corresponding circuitry. The present invention has been disclosed in the above preferred embodiments, and is not intended to limit the scope of the present invention, and those skilled in the art can make some modifications and retouchings without departing from the spirit and scope of the present invention. The protected area of the present invention is subject to the definition of the scope of the patent application.

CNTR2419I00-TW/0608-A41899-TWF 201013836 【圖式簡單說明】 第1圖係顯示根據本發明一實施例之一基底偏壓電 路’上述基底偏壓電路包括整合於P型基底上之傳統 CMOS裝置以及更顯示根據一實施例之整合於積體電路 之基底偏壓電路之示意圖。 第2圖係顯示根據本發明一實施例之整合於微處理器 晶片之基底偏壓電路之區塊圖,上述微處理器包括分佈的 φ 箝位裝置。 第3圖係顯示根據本發明一實施例所述之p型位準移 位電路之示意圖,上述P型位準移位電路可作為第1圖及 第2圖之P型位準移位電路。 第4圖係顯示根據本發明一實施例所述之n型位準移 位電路之示意圖,上述N型位準移位電路可作為第1圖及 第2圖之1ST型位準移位電路。 第5及6圖係顯示根據本發明一實施例所述之p型與 • N型緩衝器之示意圖。 第7圖係顯示根據本發明一實施例所述之整合於微處 理器晶片之選擇區域之基底偏壓電路之示意圖,上述微處 理器包括分佈的箝位裝置。 第8圖係顯示根據本發明一實施例所述之微處理器分 為多個區域之區塊圖,上述區域分別包括基底偏壓電路與 分佈的箝位裝置。 【主要元件符號說明】 CNTR2419I00-TW/0608-A41899-TWF 37 201013836 100〜積體電路 101〜P型基底 102、 202、706、810、816、822〜基底偏壓電路 103、 105、107〜N型井區 104、 106、708、710〜基底偏壓導線 109〜P型井區 111、 726、824、PI、P2、P3、P4〜P 型通道裝置 112、 718〜偏壓產生器 113、 728、826、Nl、N2、N3、N4〜N 型通道裝置 114〜控制裝置 115、117、127〜P型擴散區 116〜P型位準移位電路、LSP 118〜N型位準移位電路、LSN 119、123、125〜N型擴散區 12卜129〜閘極絕緣層 200、 700、800〜微處理器 201、 722〜P型緩衝器; 203〜N型緩衝器 301、401〜反相器 701、702、703、704〜象限 705〜其他裝置 712〜多個P型通道箝位裝置 714〜多個N型通道箝位裝置 707、802〜中央控制裝置CNTR2419I00-TW/0608-A41899-TWF 201013836 [Simplified Schematic] FIG. 1 is a diagram showing a conventional substrate bias circuit including a conventional integrated circuit on a P-type substrate according to an embodiment of the present invention. A CMOS device and a schematic diagram showing a substrate bias circuit integrated in an integrated circuit in accordance with an embodiment. Figure 2 is a block diagram showing a substrate biasing circuit integrated into a microprocessor die, including a distributed φ clamping device, in accordance with an embodiment of the present invention. 3 is a schematic diagram showing a p-type level shifting circuit according to an embodiment of the present invention, and the P-type level shifting circuit can be used as a P-type level shifting circuit of FIGS. 1 and 2. Fig. 4 is a view showing an n-type level shifting circuit according to an embodiment of the present invention, wherein the N-type level shifting circuit can be used as the 1ST type level shifting circuit of Figs. 1 and 2; 5 and 6 are schematic views showing p-type and N-type buffers according to an embodiment of the present invention. Figure 7 is a schematic illustration of a substrate biasing circuit integrated into a selected region of a microprocessor wafer, including a distributed clamping device, in accordance with an embodiment of the present invention. Figure 8 is a block diagram showing a microprocessor divided into a plurality of regions, each of which includes a substrate biasing circuit and a distributed clamping device, respectively, in accordance with an embodiment of the present invention. [Description of main component symbols] CNTR2419I00-TW/0608-A41899-TWF 37 201013836 100 to integrated circuit 101 to P type substrate 102, 202, 706, 810, 816, 822 to base bias circuit 103, 105, 107~ N-type well regions 104, 106, 708, 710 to base biasing conductors 109 to P-type well regions 111, 726, 824, PI, P2, P3, P4 to P-type channel devices 112, 718 to bias generator 113, 728, 826, N1, N2, N3, N4~N type channel device 114~ control device 115, 117, 127~P type diffusion region 116~P type level shift circuit, LSP 118~N type level shift circuit , LSN 119, 123, 125~N type diffusion region 12 129 ~ gate insulating layer 200, 700, 800~ microprocessor 201, 722~P type buffer; 203~N type buffer 301, 401~ reverse 701, 702, 703, 704 to quadrant 705 to other devices 712 to a plurality of P-channel clamp devices 714 to a plurality of N-channel clamp devices 707, 802 to central control devices

CNTR2419I00-TW/0608-A41899-TWF 201013836 716〜象限控制裝置 720〜位準移位電路 804、806、808〜區域 BCTL〜偏壓控制信號 CCTL、QCTL、CTL1、CTL2、CTL3〜控制信號 ENP、ENN〜箝位致能信號 NC1〜NC8〜N型通道箝位裝置 PEN、NEN〜箝制移位致能信號 PC1〜PC8〜P型通道箝位裝置 VBPA、VBNA〜基底偏壓 VDD〜核心電壓 VSS〜參考電壓CNTR2419I00-TW/0608-A41899-TWF 201013836 716~quad control device 720~level shift circuit 804, 806, 808~ area BCTL~bias control signal CCTL, QCTL, CTL1, CTL2, CTL3~ control signal ENP, ENN ~ Clamp enable signal NC1 ~ NC8 ~ N type channel clamp device PEN, NEN ~ clamp shift enable signal PC1 ~ PC8 ~ P type channel clamp device VBPA, VBNA ~ base bias VDD ~ core voltage VSS ~ reference Voltage

CNTR2419I00-TW/0608-A41899-TWF 39CNTR2419I00-TW/0608-A41899-TWF 39

Claims (1)

201013836 七、申請專利範園·· ,包括’· 於一第一操作模式期間 提供 .一種微處理器裝置 一第一基底偏壓導線, 一第一基底偏壓; 至小 原供應節點’提供一核心電壓; ^ ^ ^敦置,耗接於上述第一基底偏壓導後盘t 达第二電源供應節點之間;以及 料線與上 制裝置,耦接於至少一上述箝位裝置,於一第_ 操作模式期間,導通 ' 第一 基底偏壓切裝置關上述第一 操作模34第—電源供應節點以及於上述第一 作模式期間’不導通至少-上述籍位裝置。 2.如申請專利範圍第1項所述之微處理器,其中: 至少一上述箝位裴置包括一半導體裝置,具有一閘 極' 輕接於上述第—電源供應節點之—源極她接於上述 第一基底偏壓之一汲極;以及 其中上述控制裝置提供一第一箝位致能信號,以控制 上述半導體裝置之上述閘極。 3.如申請專利範圍第2項所述之微處理器’更包括: 一位準移位電路,具有接收上述第一箝位致能信號之 一輸入端,以及提供一箝制移位致能信號至上述半導體裝 置之一輸出端;以及 其申於上述第一操作模式期間,上述控制裝置設置上 述第一箝位致能信號至上述核心電壓,並導致上述位準移 位電路設置上述箝制移位致能信號至上述第一基底偏壓 40 CNTR2419I00-TW/0608-A41899-TWF 201013836 以不導通上述半導體裝置。 4.如申請專利範圍第2項所述之微處理器,其中上述 半導體裝置包括選自-P型通道裝置與—N型通道裝置之 一者。 5·如申請專利範圍第1項所述之微處理器,更包括: 一第二基底偏壓導線,於上述第一操作模式期間提供 一第二基底偏壓; 八 一第二電源供應節點,提供〆參考電壓; 其中於上述第一操作模式時,上述第一基底偏壓相對 於上述核心電壓具有—正電歷偏務,上述第二基底偏壓相 對於上述參考電壓具有一負電壓偏移; · 其中至少一上述箝位裝置包括耦接於上述第一基底 偏壓導線與上述第一電源供應節點之間之複數p型通道裝 置,以及耦接於上述第二基底偏壓導線與上述第二電源供 應郎點之間之複數N型通道裝置;以及 '、 其中上述控制裝置包括一第一輸出端以及一第二輸 出端’上述第-輸出端用以提供—第—箝位致能信號 制上述P型通道裝置,上述第二輸出端用以提供一第二— 位致能信號以控制上述N型通道裝置。 6.如申請專利範圍第5項所述之微處理器,更包括· 一 P型位準移位電路’具有耦接於上述控制裝^之上 述第-輸出端之-輸入端’與耦接於至少一上述 裝置之閘極之一輸出端;以及 、 - N型位準移位電路,具有鍵於上述控制裝置之上 CNTR2419I00-TW/0608-A41899-TWF 201013836 端之一輸八端,與轉接於至少-上述N型通道 裝置之閘極之一輸出端; i通道 其中上述控制裝置切換上一 信號於上述參考電·與上述== 間,其令上迷Ρ型位準移位電路根據上碰之 準移位電路之上述輸出端:== 壓與上述第-基底縫之間,以及上述㈣ = 根據上述第二箝位致能信號切換上述N型位準移位201013836 VII. Application for Patent Park··, including '· Provided during a first mode of operation. A microprocessor device-first substrate biasing wire, a first substrate bias; to the small original supply node' provides a core a voltage; ^ ^ ^, disposed between the first substrate biasing and guiding the rear disk t to the second power supply node; and the material line and the upper device coupled to the at least one of the clamping devices, During the first mode of operation, the first substrate biasing device is turned off to turn off the first operating mode 34 first power supply node and during the first mode of operation 'not conducting at least the above-described home device. 2. The microprocessor of claim 1, wherein: at least one of the clamp devices comprises a semiconductor device having a gate connected to the first power supply node - a source Depressing one of the first substrate biases; and wherein the control device provides a first clamp enable signal to control the gate of the semiconductor device. 3. The microprocessor of claim 2, further comprising: a quasi-shift circuit having an input for receiving the first clamp enable signal and providing a clamp shift enable signal And an output terminal of the semiconductor device; and during the first operation mode, the control device sets the first clamp enable signal to the core voltage, and causes the level shift circuit to set the clamp shift The enable signal is applied to the first substrate bias 40 CNTR2419I00-TW/0608-A41899-TWF 201013836 to disable the semiconductor device. 4. The microprocessor of claim 2, wherein the semiconductor device comprises one selected from the group consisting of a -P type channel device and an -N type channel device. 5. The microprocessor of claim 1, further comprising: a second substrate biasing wire, providing a second substrate bias during the first mode of operation; and an eighth power supply node, Providing a 〆 reference voltage; wherein, in the first mode of operation, the first substrate bias has a positive-element bias with respect to the core voltage, and the second substrate bias has a negative voltage offset with respect to the reference voltage At least one of the above-mentioned clamping devices includes a plurality of p-type channel devices coupled between the first substrate biasing wires and the first power supply node, and coupled to the second substrate biasing wires and the first a plurality of N-channel devices between the power supply points; and ', wherein the control device includes a first output terminal and a second output terminal, the first output terminal is configured to provide a --clamp enable signal The P-channel device is configured to provide a second bit enable signal to control the N-channel device. 6. The microprocessor of claim 5, further comprising: a P-type level shifting circuit 'having an input-portion coupled to the first output terminal of the control device And at least one output terminal of the gate of the device; and, an N-type level shifting circuit, having a button on the CNTR2419I00-TW/0608-A41899-TWF 201013836 end of the control device, and Translating at least one of the gates of the gate of the N-channel device; wherein the control device switches the previous signal between the reference power and the above ==, which causes the upper level shift circuit And the output terminal of the shift circuit according to the upper touch: between the == voltage and the first base-stitch, and (4) above: switching the N-type shift according to the second clamp enable signal 之上7述於上述核心電壓與上述第二基底偏壓之間。 ’ β專利範圍第6項所述之微處理H,更包括· 上述;具键於上述ρ型位準移位電路之 上达輸出端之一輸入端’以及耦接於至少一上述 裝置之一輸出端; 一 Ν型緩衝器,具有耦接於上述Ν型位準移位電路之 上述輸出端之一輸入端,以及耦接於至少一上述ν型通道 裝置之一輸出端;以及The above 7 is between the above core voltage and the second substrate bias. The micro-processing H described in item 6 of the β patent range further includes: the above-mentioned p-type level shifting circuit having one input end of the output end and one of the at least one of the above devices An output terminal having an input terminal coupled to the output terminal of the Ν-type level shifting circuit and coupled to an output of at least one of the ν-type channel devices; 其中上述Ρ型緩衝器切換上述ρ型緩衝器之上述輸出 端於上述參考電壓與上述第一基底偏壓之間,上述Ν型緩 衝器切換上述Ν型緩衝器之上述輸出端於上述核心電壓 與上述第二基底偏壓之間。 8.如申請專利範圍第1項所述之微處理器,更包括 一基底,具有一第一區域與一第二區域; 複數第一半導體裝置,位於上述第一區域; 複數第二半導體裝置,位於上述第二區域;以及 CNTR2419IO0-TW/0608-A41899-TWF 42 201013836 、、其中上述第一基底偏壓導線繞線於位於上述基底之 上述第一區域之上述第一半導體裝置以於上述第一操作 模式偏壓上述第一半導體裝置,而上述第二半導體裝置保 持電力開啟。 味9.如申睛專利範圍第8項所述之微處理器,其中上述 柑位裝置係沿著位於上述基底之上述第一區域之上述第 一基底偏壓導線分佈。 1〇’如申請專利範圍第1項所述之微處理器,更包括: 基底,具有一第一區域與一第二區域; 其中上述第一基底偏壓導線位於上述第一區域; 一第二基底偏壓導線位於上述第二區域,並 三操作模式提供—第二基底偏壓; 、 於上包括餘上述第—區域並麵接 . -^^ ^ :上述第一區域並耦接於上 其中上述控制裝置於上述第_ 述第一箝位裝置與上述第二箝導通上 底偏壓導線與上述第二基底 柑彳上述第一基 ㈣點,於上述第i作模式不電源供 置且導通上述第二箝位裝置,u等通上迷第一箝位裝 不導通上述第二箝位裝置。、;述第三操作模式期間 lh如申請專利範圍第】項所述之微處理器,更包括: CNTR2419I00-TW/0608-A41899-TWF 43 201013836 一基底’具有-第-區域與-第二區域; f中上述第—基底偏科線位於上述第-區域; 第導線位於上述第二區域,並且於上述 第一操作模式提供一第二基底偏屢; 於二二一上述箝位裝置包括位於上述第-區域並輕接 =-基底偏物與上述第一電源供應節點= 參 :第第=底辑線與上述第-電源供應節點二 數第一箝位裝置;以及 其中上述控制裝置於上述第二操作模式期間,導通上 =第=位裝置且不導通上述第二箝位裝置以箝制上述 第基底偏壓導線至上述第一電源供應節點,於上述第一 ,作模式期間不導通上述第—箝位裝置且導通上述第二 箝位裝置以箝制上述第二基底偏壓導線至上述第 供應節點。 ' !2. —種積體電路,包括: 一基底; 一第一基底偏壓導線與一第二基底偏壓導線,位於上 述基底; 一第一電源供應導體,位於上述基底,用以提供一核 心電壓,上述核心電壓係相對於位於上述基底之一第二^ 源供應導體所提供之一參考電壓; 其中於上述積體電路之一第一操作模式期間,提供一 第一基底偏壓於上述第一基底偏壓導線,且提供—第一美 CNTR2419IO0-TW/0608-A41899-TWF 44 201013836 底偏壓於上述第二基底偏壓導線,其中上述第一基底偏壓 高於上述核心電壓而上述第二基底偏壓低於上述參考電 壓; 至少一第一箝位裝置提供於上述基底,至少一上述第 一箝位裝置分別耦接於上述第一電源供應導體與上述第 一基底偏壓導線之間; 至少一第一·箝位裝置提供於上述基底,至少一上述第The Ρ type buffer switches the output end of the p-type buffer between the reference voltage and the first substrate bias, and the 缓冲器 type buffer switches the output end of the 缓冲器 type buffer to the core voltage and Between the above second substrate biases. 8. The microprocessor of claim 1, further comprising a substrate having a first region and a second region; a plurality of first semiconductor devices located in the first region; and a plurality of second semiconductor devices, Located in the second region; and CNTR2419IO0-TW/0608-A41899-TWF 42 201013836, wherein the first substrate bias wire is wound around the first semiconductor device located in the first region of the substrate to be the first The operating mode biases the first semiconductor device while the second semiconductor device remains powered on. The microprocessor of claim 8, wherein the civic device is distributed along the first substrate biasing wire located in the first region of the substrate. The microprocessor of claim 1, further comprising: a substrate having a first region and a second region; wherein the first substrate biasing wire is located in the first region; The substrate biasing wire is located in the second region, and provides a second substrate bias in a three-operation mode; and includes the remaining first region and is face-to-face. -^^ ^: the first region is coupled to the upper portion The control device is configured to connect the upper clamp biasing lead and the second base cilia to the first base (four) point in the first clamp device and the second clamp, and the power supply is not provided and turned on in the ith mode In the second clamping device, the first clamping device does not conduct the second clamping device. The microprocessor described in the third mode of operation lh, as described in the scope of the patent application, includes: CNTR2419I00-TW/0608-A41899-TWF 43 201013836 A substrate 'having a - region-and a second region The above-mentioned first-substrate bias line is located in the above-mentioned first region; the first wire is located in the second region, and a second substrate is provided in the first operation mode; and the clamping device comprises the above-mentioned clamping device a first region and a light connection = - a base bias and the first power supply node = a reference: a first = bottom line and the first power supply node two first clamp devices; and wherein the control device is in the above During the second mode of operation, turning on the upper = the third bit device and not conducting the second clamping device to clamp the first base bias wire to the first power supply node, during the first mode, the first mode is not turned on And clamping the second clamping device to clamp the second substrate biasing wire to the first supply node. An integrated circuit comprising: a substrate; a first substrate biasing wire and a second substrate biasing wire on the substrate; a first power supply conductor located on the substrate to provide a a core voltage, wherein the core voltage is relative to a reference voltage provided by a second source supply conductor of the substrate; wherein during the first mode of operation of the integrated circuit, a first substrate bias is provided The first substrate biases the wire and provides - the first US CNTR2419IO0-TW/0608-A41899-TWF 44 201013836 bottom biased to the second substrate bias wire, wherein the first substrate bias is higher than the core voltage The second substrate bias is lower than the reference voltage; at least one first clamping device is provided on the substrate, and at least one of the first clamping devices is respectively coupled between the first power supply conductor and the first substrate bias wire At least one first clamping device is provided on the substrate, at least one of the above 二箝位裝置分別耦接於上述第二電源供應導體與上述第 二基底偏壓導線之間;以及 一控制裝置具有用以控制至少一上述第一箝位裝置 之一第一輸出端,以及用以控制至少一上述第二箝位裝置 之一第二輸出端; 其中上述控制裝置於上述第一操作模式不導通至少 -上述第-箝位裝置與至少—上述第二箝位裝置,於上述 第二操作模式導通至少一上述第一箝位裝置與至少一上 述第二箝位裝置,以_上述第—基底偏壓導線至上述第 -電源供應導體以及㈣彳±述第二基底偏壓導線至上述 第二電源供應導體。 13.如申請專利範圍帛12項所述之積體電路,其中至 少-上述第-箝位裝置包括—第”型通道裝置,具 接於上述第-電源供應導體之—源極、㈣於上述第 底偏壓導線之-汲極及由上述控制装置之上 端所控制之-閘極’以及其中至少一上述第 二 括一第一 N型通道裝置,具有耦接 、匕 上述第二電源供應導 CNTE12419I00-TW/Q608-A41899-TWF 45 201013836 體之-源極、叙接於上述第二基底偏塵導線之 上述控制裝置之上述第二輪出端所控制之,極及由 ,如申請專利範圍第13項所述之積體= 述通道裝置包括_於上述第—基路=上 -基底接點以及其t上述第_ N型通 ,導線之 上述第二基底偏壓導線之一基底接點。 匕括耦接於 括,.如申請專利範圍第13項所述之積體電路,更包 -第-位準移位電路’具有_於上述控制 述第-輸出端之-輸入端’以及稱接於上述第上 裝置之上述_之-輸出端,其中上述控制 =通道 控制裝置之上述第-輸出端至上述參考電壓通^述 第-P型通道裝置’及切換上述控制裝置之上述 端至上述核心電壓以不導通上述第—?型通輸出 其中上述第-位準移位電路切換上述第一p型通及 上述閘極至上述參考電壓以導通上述第一二之 及切換上述第-P型通道裝置之上述閘極至上述d ’ 偏壓以不導通上述第一 P型通道裝置;以及 基底 -第二位準移位電路,具有㈣於上述控制裝 述第二輸出端之一輸入端,以及耦接於上述第一 N 裝置之上述閘極之一輸出端,其中上述控制裝 ,道 控制裝置之上述第二輸出端至上述核心電壓以返 第一 N型通道裝置,及切換上述控制裝置之上述迷 端至上述參考電壓以不導通上述第一N型通道裳一輸出 CNTR2419I00-TW/0608-A41899-TWF 201013836 • 其中上述第二位準移位電路切換上述第一 N型通道裝置 一 之上述閘極至上述核心電壓以導通將上述第一 N型通道 裝置,及切換上述第一 N型通道裝置之上述閘極至第二基 底偏壓以不導通上述第一 N型通道裝置。 16. 如申請專利範圍第15項所述之積體電路,更包 括: 至少一上述第一箝位裝置包括一第二P型通道裝置, 具有耦接於上述第一電源供應導體之一源極,耦接於上述 • 第一基底偏壓導線之一汲極與一閘極; 一第一緩衝器,具有耦接於上述第一位準移位電路之 上述輸出端之一輸入端,以及耦接上述第二P型通道裝置 之上述閘極之一輸出端,其中上述第一缓衝器切換上述第 一緩衝器之上述輸出端隨著上述第一位準移位電路之上 述輸出端於上述參考電壓與上述第一基底偏壓之間; 至少一上述第二箝位裝置包括一第二N型通道裝置, 具有耦接於上述第二電源供應導體之一源極、耦接於上述 論 第二基底偏壓導線之一汲極與一閘極;以及 一第二緩衝器,具有耦接於上述第二位準移位電路之 上述輸出端之一輸入端,以及耦接上述第二N型通道裝置 之上述閘極之一輸出端,其中上述第二緩衝器切換上述第 二緩衝器之上述輸出端隨著上述第二位準移位電路之上 述輸出端於上述核心電壓與上述第二基底偏壓之間。 17. 如申請專利範圍第12項所述之積體電路,其中至 少一上述第一箝位裝置包括沿著上述第一基底偏壓導線 CNTR2419I00-TW/0608-A41899-TWF 47 201013836 分佈的複數第一箝位裝置,用以於上述第二狀態導通上述 第一箝位裝置以維持上述第一基底偏壓導線之電壓於相 對於上述核心電壓之變動在一第一既定最小電壓位準 内,其中至少一上述第二箝位裝置包括沿著上述第二基底 偏壓導線分佈的複數第二箝位裝置,用以於上述第二操作 模式導通上述第二箝位裝置以維持上述第二基底偏壓導 線之電壓於相對於上述參考電壓之變動在一第二既定最 小電壓位準内。 18. 如申請專利範圍第12項所述之積體電路,其中上 述基底分為第一區域與第二區域,分別具有複數半導體裝 置,以及其中上述第一基底偏壓導線、第二基底偏壓導線 與至少一上述第一箝位裝置位於上述基底之上述第一區 域。 19. 一種晶片雜訊減少方法,適用於一微處理器晶 片,上述微處理器晶片包括用以減少次臨界漏電流之一第 一基底偏壓導線,上述方法包括: 當上述微處理器晶片於一第一電力狀態,箝制上述第 一基底偏壓導線至一核心電壓;以及 當上述微處理器晶片於一第二電力狀態,不箝制上述 第一基底偏壓導線,並且提供一第一基底偏壓至上述第一 基底偏壓導線。 20. 如申請專利範圍第19項所述之晶片雜訊減少方 法,其中箝制上述第一基底偏壓導線至上述核心電壓之步 驟包括導通所選取的複數第一箝位裝置,上述複數第一箝 CNTR2419I00-TW/0608-A41899-TWF 48 201013836 - 位裝置用來維持上述基底偏壓導線之電壓於相對於上述 , 核心電壓之變動在一第一既定最小電壓位準且複數第一 箝位裝置沿著上述第一基底偏壓導線分佈。 21. 如申請專利範圍第19項所述之晶片雜訊減少方 法,更包括: 耦接一第一半導體裝置之一汲極與一源極於上述第 一基底偏壓導線與上述核心電壓之間; 當微處理器晶片於上述第一電力狀態,導通上述第一 • 半導體裝置; 當微處理器微處理器晶片於上述第二電力狀態,不導 通上述第一半導體裝置。 22. 如申請專利範圍第21項所述之晶片雜訊減少方 法,其中: 提供第一基底偏壓之步驟更包括提供一偏移電壓以 驅動上述第一基底偏壓導線的電壓高於上述核心電壓;以 及 其中不導通上述第一半導體裝置之步驟包括提供一 第一箝制致能信號,上述第一箝制致能信號設置上述第一 半導體裝置之一閘極至高於上述核心電壓上述偏移電壓 的電壓位準。 23. 如申請專利範圍第21項所述之晶片雜訊減少方 法,其中: 提供第一基底偏壓之步驟更包括提供一偏移電壓以 驅凍上述第一基底偏壓導線的電壓低於上述核心電壓;以 CNTR2419I00-TW/0608-A41899-TWF 49 201013836 及 其中不導通上述第一半導體裝置之步驟包括提供一 第一箝制致能信號,上述第一箝制致能信號設置上述第一 半導體裝置之一閘極至低於上述核心電壓上述偏移電壓 的電壓位準。 24. 如申請專利範圍第22項所述之晶片雜訊減少方 法,更包括: 耦接一第二半導體裝置之一汲極與一源極於上述第 一基底偏壓導線與上述核心電壓之間;以及 於上述微處理器晶片提供一緩衝器,上述緩衝器用以 緩衝上述第一箝位致能信號以提供一緩衝箝位致能信號 至上述第二半導體裝置之一閘極,其中上述緩衝箝位致能 信號與上述第一箝位致能信號之電壓位準相同。 25. 如申請專利範圍第23項所述之晶片雜訊減少方 法,更包括: 耦接一第二半導體裝置之一汲極與一源極於上述第 一基底偏壓導線與上述核心電壓之間;以及 於上述微處理器晶片提供一缓衝器,上述緩衝器用以 缓衝上述第一箝位致能信號以提供一緩衝箝位致能信號 至上述第二半導體裝置之一閘極,其中上述缓衝箝位致能 信號與上述第一箝位致能信號之電壓位準相同。 26. 如申請專利範圍第19項所述之晶片雜訊減少方 法,其中上述微處理器晶片分為一第一區域與一第二區 域,並且包括一第二基底偏壓導線,其中上述第一基底偏 CNTR2419IO0-TW/06O8-A41899-TWF 50 201013836 . 壓導線位於上述第一區域,上述第二基底偏壓導線位於上 述第二區域,上述晶片雜訊減少方法更包括: 當上述微處理器晶片於上述第一電力狀態與上述第 二電力狀態時,箝制上述第二基底偏壓導線至上述核心電 壓;以及 當上述微處理器晶片於一第三電力狀態,不箝制上述 第二基底偏壓導線並提供一第二基底偏壓至上述第二基 底偏壓導線。 ❹ 27.如申請專利範圍第19項所述之晶片雜訊減少方 法,其中上述微處理器晶片分為一第一區域與一第二區 域,並且包括一第二基底偏壓導線,其中上述第一基底偏 壓導線位於上述第一區域,上述第二基底偏壓導線位於上 述第二區域,上述晶片雜訊減少方法更包括: 當上述微處理器晶片於上述第二電力狀態時,箝制上 述第二基底偏壓導線至上述核心電壓;以及 當上述微處理器晶片於上述第一電力狀態,不箝制上 — 述第二基底偏壓導線並提供一第二基底偏壓至上述第二 基底偏壓導線。 CNTR2419IO0-TW/O608-A41899-TWF 51The second clamping device is respectively coupled between the second power supply conductor and the second substrate biasing wire; and a control device has a first output for controlling at least one of the first clamping devices, and Controlling at least one of the second clamping devices of the second clamping device; wherein the control device does not conduct at least the first clamping device and the at least the second clamping device in the first operating mode, The second operation mode turns on at least one of the first clamping device and the at least one second clamping device to: the first substrate biasing wire to the first power supply conductor and (4) the second substrate bias wire to The second power supply conductor described above. 13. The integrated circuit of claim 12, wherein at least the first clamping device comprises a -" channel device having a source connected to the first power supply conductor, (d) a second-type power supply guide for the bottom-biased conductor-drain and the gate-controlled by the upper end of the control device and at least one of the second and first first-type N-channel devices CNTE12419I00-TW/Q608-A41899-TWF 45 201013836 The source-source, which is controlled by the above-mentioned second round of the above-mentioned control device of the second base dust guide, is controlled by the patent application scope The integrated body according to Item 13 includes: the above-mentioned first-base=upper-substrate contact and t-the above-mentioned N-th-pass, one of the second base biasing wires of the wire In addition, the integrated circuit described in claim 13 of the patent application, the package-first level shifting circuit 'has the input-output terminal of the above-mentioned control-output terminal and Said to be connected to the above-mentioned upper device The control/the above-mentioned first output terminal of the channel control device to the reference voltage to the first-P-type channel device' and the end of the control device to the core voltage to not conduct the first-type output The first-level shifting circuit switches the first p-type pass and the gate to the reference voltage to turn on the first two and switch the gate of the first-P-type channel device to the d' bias The first P-channel device is not turned on; and the substrate-second level shift circuit has (4) an input terminal of the second output end of the control device, and the gate coupled to the first N device One of the output terminals, wherein the control device, the second output end of the track control device to the core voltage is returned to the first N-type channel device, and the fan terminal of the control device is switched to the reference voltage to not conduct the above The first N-channel output is CNTR2419I00-TW/0608-A41899-TWF 201013836. The second level shift circuit is switched over the first N-channel device. Passing the gate to the core voltage to turn on the first N-type channel device, and switching the gate of the first N-type channel device to the second substrate to not turn on the first N-type channel device. The integrated circuit of claim 15 further comprising: at least one of the first clamping devices comprising a second P-channel device having a source coupled to the first power supply conductor and coupled The first buffer has a drain and a gate; and the first buffer has an input coupled to the output of the first level shifting circuit, and coupled to the first An output terminal of the gate of the second P-type channel device, wherein the first buffer switches the output end of the first buffer with the output terminal of the first level shifting circuit at the reference voltage Between the first substrate biases; at least one of the second clamping devices includes a second N-channel device coupled to one of the second power supply conductors and coupled to the second base One of the biasing wires and one of the gates; and a second buffer having an input coupled to the output of the second level shifting circuit and coupled to the second N-channel device One of the output terminals of the gate, wherein the second buffer switches the output end of the second buffer, and the output terminal of the second level shifting circuit is biased to the core substrate and the second substrate between. 17. The integrated circuit of claim 12, wherein at least one of the first clamping devices comprises a plurality of first base biasing wires CNTR2419I00-TW/0608-A41899-TWF 47 201013836. a clamping device for conducting the first clamping device in the second state to maintain the voltage of the first substrate biasing wire within a first predetermined minimum voltage level with respect to the variation of the core voltage, wherein At least one of the second clamping devices includes a plurality of second clamping devices distributed along the second substrate biasing conductor for conducting the second clamping device in the second operating mode to maintain the second substrate bias The voltage of the wire is within a second predetermined minimum voltage level with respect to the change in the reference voltage. 18. The integrated circuit of claim 12, wherein the substrate is divided into a first region and a second region, each having a plurality of semiconductor devices, and wherein the first substrate bias wire and the second substrate are biased A wire and at least one of the first clamping devices are located in the first region of the substrate. 19. A method of reducing wafer noise for a microprocessor chip, the microprocessor chip comprising a first substrate biasing wire for reducing a sub-critical leakage current, the method comprising: when the microprocessor chip is a first power state, clamping the first substrate bias wire to a core voltage; and when the microprocessor chip is in a second power state, not clamping the first substrate bias wire and providing a first substrate bias Pressing to the first substrate biasing wire described above. 20. The wafer noise reduction method of claim 19, wherein the step of clamping the first substrate bias wire to the core voltage comprises turning on the selected plurality of first clamping devices, the plurality of first clamps CNTR2419I00-TW/0608-A41899-TWF 48 201013836 - The bit device is used to maintain the voltage of the substrate biasing conductor above the core voltage variation relative to the above, at a first predetermined minimum voltage level and along the plurality of first clamping devices The first substrate biased conductor is distributed as described above. 21. The wafer noise reduction method of claim 19, further comprising: coupling a drain of a first semiconductor device and a source between the first substrate bias wire and the core voltage When the microprocessor chip is in the first power state, turning on the first semiconductor device; when the microprocessor microprocessor is in the second power state, the first semiconductor device is not turned on. 22. The wafer noise reduction method of claim 21, wherein: the step of providing a first substrate bias further comprises providing an offset voltage to drive the voltage of the first substrate biasing conductor to be higher than the core And the step of not conducting the first semiconductor device includes: providing a first clamp enable signal, wherein the first clamp enable signal sets a gate of the first semiconductor device to be higher than the offset voltage of the core voltage Voltage level. 23. The wafer noise reduction method of claim 21, wherein: the step of providing a first substrate bias further comprises providing an offset voltage to drive the voltage of the first substrate bias wire to be lower than the above Core voltage; CNTR2419I00-TW/0608-A41899-TWF 49 201013836 and the step of not conducting the first semiconductor device therein includes providing a first clamp enable signal, and the first clamp enable signal is disposed on the first semiconductor device A gate to a voltage level lower than the above-mentioned offset voltage of the core voltage. 24. The wafer noise reduction method of claim 22, further comprising: coupling a drain of a second semiconductor device and a source between the first substrate bias conductor and the core voltage And providing a buffer to the microprocessor chip, wherein the buffer is configured to buffer the first clamp enable signal to provide a buffer clamp enable signal to one of the gates of the second semiconductor device, wherein the buffer clamp The bit enable signal is the same as the voltage level of the first clamp enable signal. 25. The wafer noise reduction method of claim 23, further comprising: coupling a drain of a second semiconductor device and a source between the first substrate bias wire and the core voltage And providing a buffer to the microprocessor chip, wherein the buffer is configured to buffer the first clamp enable signal to provide a buffer clamp enable signal to one of the gates of the second semiconductor device, wherein The buffer clamp enable signal is the same as the voltage level of the first clamp enable signal. 26. The wafer noise reduction method of claim 19, wherein the microprocessor chip is divided into a first region and a second region, and includes a second substrate bias wire, wherein the first The substrate is biased CNTR2419IO0-TW/06O8-A41899-TWF 50 201013836. The pressure wire is located in the first region, and the second substrate bias wire is located in the second region, and the chip noise reduction method further comprises: when the microprocessor chip And clamping the second base bias wire to the core voltage during the first power state and the second power state; and clamping the second base bias wire when the microprocessor chip is in a third power state And providing a second substrate bias to the second substrate biasing wire. The wafer noise reduction method of claim 19, wherein the microprocessor chip is divided into a first region and a second region, and includes a second substrate bias wire, wherein the a substrate biasing wire is located in the first region, and the second substrate biasing wire is located in the second region, and the chip noise reduction method further comprises: clamping the first chip when the microprocessor chip is in the second power state a second substrate biasing wire to the core voltage; and when the microprocessor chip is in the first power state, not clamping - the second substrate biasing wire and providing a second substrate bias to the second substrate bias wire. CNTR2419IO0-TW/O608-A41899-TWF 51
TW98132275A 2008-09-25 2009-09-24 Microprocessors﹑intergarated circuits and methods for reducing noises thereof TWI405297B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US12/237,463 US7920019B2 (en) 2008-09-25 2008-09-25 Microprocessor with substrate bias clamps
US12/237,483 US7978001B2 (en) 2008-09-25 2008-09-25 Microprocessor with selective substrate biasing for clock-gated functional blocks

Publications (2)

Publication Number Publication Date
TW201013836A true TW201013836A (en) 2010-04-01
TWI405297B TWI405297B (en) 2013-08-11

Family

ID=42049054

Family Applications (2)

Application Number Title Priority Date Filing Date
TW98132275A TWI405297B (en) 2008-09-25 2009-09-24 Microprocessors﹑intergarated circuits and methods for reducing noises thereof
TW98132273A TWI388977B (en) 2008-09-25 2009-09-24 Microprocessors, intergarated circuits and methods for selectively biasing substrates

Family Applications After (1)

Application Number Title Priority Date Filing Date
TW98132273A TWI388977B (en) 2008-09-25 2009-09-24 Microprocessors, intergarated circuits and methods for selectively biasing substrates

Country Status (2)

Country Link
CN (4) CN105099428B (en)
TW (2) TWI405297B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101924456B (en) * 2010-08-04 2012-11-07 钰创科技股份有限公司 Buffering drive circuit, buffer and performance improvement method thereof
TW201250449A (en) * 2010-12-12 2012-12-16 Via Tech Inc An apparatus and a method for dynamically varying a bias voltage applied to a substrate of an integrated circuit
CN103178830B (en) * 2011-12-23 2016-03-23 联芯科技有限公司 substrate selection circuit
TWI569126B (en) 2015-08-31 2017-02-01 威盛電子股份有限公司 Output buffer apparatus

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR0169157B1 (en) * 1993-11-29 1999-02-01 기다오까 다까시 Semiconductor circuit and mos-dram
JP3633061B2 (en) * 1995-10-19 2005-03-30 三菱電機株式会社 Semiconductor integrated circuit device
TW400650B (en) * 1996-11-26 2000-08-01 Hitachi Ltd Semiconductor integrated circuit device
US6218895B1 (en) * 1997-06-20 2001-04-17 Intel Corporation Multiple well transistor circuits having forward body bias
WO1998059419A1 (en) * 1997-06-20 1998-12-30 Intel Corporation Forward body bias transistor circuits
TW453032B (en) * 1998-09-09 2001-09-01 Hitachi Ltd Semiconductor integrated circuit apparatus
JP2002064150A (en) * 2000-06-05 2002-02-28 Mitsubishi Electric Corp Semiconductor device
CN101253686B (en) * 2005-09-02 2010-12-29 松下电器产业株式会社 Semiconductor integrated circuit
CN1992269A (en) * 2005-12-28 2007-07-04 松下电器产业株式会社 Semiconductor integrated circuit device
US7355437B2 (en) * 2006-03-06 2008-04-08 Altera Corporation Latch-up prevention circuitry for integrated circuits with transistor body biasing
JP2008182004A (en) * 2007-01-24 2008-08-07 Renesas Technology Corp Semiconductor integrated circuit

Also Published As

Publication number Publication date
CN101686049B (en) 2012-05-23
CN102647175B (en) 2016-04-13
CN105099428A (en) 2015-11-25
CN101686049A (en) 2010-03-31
TWI405297B (en) 2013-08-11
CN101686050A (en) 2010-03-31
TW201013391A (en) 2010-04-01
CN105099428B (en) 2020-06-12
TWI388977B (en) 2013-03-11
CN102647175A (en) 2012-08-22

Similar Documents

Publication Publication Date Title
Lorenzo et al. Review of circuit level leakage minimization techniques in CMOS VLSI circuits
US7425845B2 (en) Semiconductor integrated circuit
TW200425136A (en) Reduced integrated circuit chip leakage and method of reducing leakage
US7703062B2 (en) Semiconductor integrated circuit and method of designing layout of the same
US6518826B2 (en) Method and apparatus for dynamic leakage control
JP2011147038A (en) Semiconductor device and data processing system including the same
TW201013836A (en) Microprocessors, intergarated circuits and methods for reducing noises thereof
EP2326008A1 (en) A simple self-adjusting overvoltage-protection circuit for low voltage CMOS input and output interface circuits with high voltage tolerance and with full rail-to-rail bidirectional voltage levels
US7978001B2 (en) Microprocessor with selective substrate biasing for clock-gated functional blocks
Chen et al. I/sub DDQ/testing for deep-submicron ICs: challenges and solutions
Li et al. Energy-efficient full-swing logic circuits with unipolar TFTs on flexible substrates
JP2007042797A (en) Semiconductor integrated circuit
US7920019B2 (en) Microprocessor with substrate bias clamps
US6801057B2 (en) Silicon-on-insulator dynamic logic
CN107452740B (en) Integrated circuit with spare unit
CN105703761A (en) I/O driving circuit
JP3804633B2 (en) Semiconductor integrated circuit
JP2004228787A (en) Semiconductor integrated circuit
JP4557046B2 (en) Output buffer circuit and integrated circuit
US6975143B2 (en) Static logic design for CMOS
US9264040B2 (en) Low leakage CMOS cell with low voltage swing
US6552566B2 (en) Logic array circuits using silicon-on-insulator logic
US8344757B2 (en) Semiconductor device and data processing system including the same
US20240039537A1 (en) High-voltage fault protection circuit
CN100533733C (en) Layout circuit with stable guiding current and IC chip with the same