CN102944803A - Leakage power estimation - Google Patents

Abstract

The present invention describes methods and apparatus for providing leakage power estimates. In one embodiment, one or more sensed temperature values (108) and one or more voltage values (110) are used to determine leakage power of integrated circuit (IC) components. Other embodiments are also described.

Description

Leakage Power Estimation

This application is a divisional application of a Chinese invention patent application with an application date of June 30, 2006, an application number of 200680054765.7, and an invention title of "Leakage Power Estimation".

Background technique

The present invention relates generally to the field of electronics, and more particularly, embodiments of the invention relate to leakage power estimation in integrated circuit (IC) devices.

technical field

Both dynamic power consumption and leakage power consumption are one of the main concerns in IC design. In particular, subthreshold leakage (or leakage power) may increase with each successive design stage. For example, when the supply voltage is reduced (eg, to reduce dynamic power consumption), the threshold voltage is also lowered (eg, to maintain low gate delay or high frequency). However, lowering the threshold voltage may affect leakage power in a non-linear fashion.

In some embodiments, the leakage power may be assumed to be constant at run time. However, leakage power can also vary during operation, for example with changes in temperature, supply voltage, or threshold voltage. Therefore, without knowing the leakage power, power management techniques can be less accurate.

Description of drawings

The detailed description will be made with reference to the accompanying drawings. In the figures, the left-most digit(s) in a reference number indicates the figure number in which the reference number first appears. In different drawings, the same reference numerals are used to designate similar or identical items.

1, 5 and 6 show block diagrams of computing systems according to various embodiments of the invention.

2A and 2B represent block diagrams of portions of a leakage power estimation system according to various embodiments.

Figure 3 shows a block diagram of a processor core according to an embodiment.

Fig. 4 represents a flowchart of a method according to an embodiment.

Detailed ways

In the following description, numerous specific details are given to provide a thorough understanding of various embodiments. However, various embodiments of the invention may be practiced without these specific details. In other instances, well-known methods, procedures, components and circuits have not been described in detail, so as not to highlight particular embodiments of the invention. The various aspects of the embodiments of the invention may be implemented by various means, such as integrated semiconductor circuits ("hardware"), computer readable instructions constituting one or more programs ("software"), or a combination of hardware and software. combination. For purposes of this specification, references to "logic" refer to hardware, software, or a combination thereof.

Some embodiments discussed herein may provide efficient techniques for estimating leakage power (eg, static or subthreshold leakage power generated by one or more components of an IC device). In embodiments, this leakage power may be caused by one or more changes, such as changes in temperature and/or voltage (eg, threshold and/or supply voltage). Moreover, some embodiments discussed herein may be applied to various computing systems, such as those discussed with reference to FIGS. 1 , 5 , and 6 . More specifically, FIG. 1 is a block diagram of a computing system 100 according to an embodiment. System 100 may include one or more domains 102 - 1 through 102 -M (collectively referred to herein as "domains 102"). Each of domains 102 - 1 through 102 -M may include a wide variety of components, but for clarity, only example components are drawn for domains 102 - 1 and 102 - 2 . Also, each domain 102 may correspond to a portion of a computing system (eg, a component discussed with reference to FIGS. 5 and 6 , or more generally, one or more transistors of an IC device). In an embodiment, each of domains 102 may include different circuitry (or logic) clocked by a clock signal that may be different than the clock signal used in the other domains. In one embodiment, one or more of these clock signals may be synchronized on average, or otherwise related (eg, have a relationship that repeats itself or does not repeat itself over time).

As shown in FIG. 1 , each domain may communicate data with other domains through one or more buffers 104 . In one embodiment, buffer 104 may be a first-in-first-out (FIFO) buffer. Each domain may include logic (such as logic 106-1 and 106-2 shown with reference to domains 102-1 and 102-2, respectively, collectively referred to herein as "logic") to estimate the leakage power of one or more components within the corresponding domain. 106"), one or more temperature sensors (such as sensors 108-1 and 108-2 shown with reference to fields 102-1 and 102-2, respectively), control frequency and/or voltage level and/or provide current threshold logic for voltage and/or supply voltage values (such as logic 110-1 and 110-2 shown with reference to domains 102-1 and 102-2, respectively), and logic to manage power consumption of one or more components within the corresponding domain ( For example, logic 112 - 1 and 112 - 2 shown with reference to fields 102 - 1 and 102 - 2 , respectively, collectively referred to herein as "logic 112 "). In an embodiment, the threshold voltage of a transistor can be adjusted by applying a current to the body (substrate) of the transistor.

In various embodiments, power management logic 112 may regulate power consumption of one or more components of the corresponding domain. For example, logic 112 may utilize information such as leakage power estimates (such as values provided by corresponding logic 106 ), dynamic power estimates, and/or some other information (such as instructions per cycle, cache errors, etc.). Also, the logic 112 may adjust the frequency of a clock signal (eg, a clock signal used within at least a portion of the respective domain). In an embodiment, logic 112 may shut down one or more components: for example, one or more processor cores or portions of these processor cores (eg, different pipelines, etc.) and/or data caches (eg, May include various levels of cache, such as first level (L1), second level (L2), or others) or portions of data caches (such as different cache clusters).

2A and 2B show block diagrams of components of leakage power estimation system 200 and system 250 according to various embodiments. In one embodiment, system 200 and system 250 are the same as or similar to logic 106 discussed with reference to FIG. 1 . In embodiments, the memory cells discussed with reference to FIGS. 2A and 2B may be the same or similar to the memory components discussed with reference to FIGS. 5 and/or 6 .

As shown in FIGS. 2A and 2B , systems 200 and 250 may include a temperature calibration factor storage unit 202 (eg, configured to store a plurality of temperature calibration factor values). Storage unit 202 may receive detected temperature values from sensors 108 corresponding to one or more components, such as those discussed with reference to FIGS. 1 , 5 , and 6 . System 200 may also include a voltage scaling factor storage unit 204 (eg, to store a plurality of voltage factor values) and a reference leakage storage unit 206 (eg, to store a reference or baseline leakage power value). The reference leakage value stored in the storage unit 206 may be determined during design (for example, through simulation or circuit measurement), or determined during testing. For example, for designs with higher variability, a baseline leakage value can be determined at test time (this is because, for each chip and/or functional block, this baseline value can be calculated individually to accommodate each circuit detail estimate).

In an embodiment, the system 200 may further include rounding logic 210 for rounding the temperature value received from the sensor 108 (eg, rounding the detected value to the closest value to the value stored in the storage unit 202 ). Interpolation logic 212 may interpolate the value output by storage unit 202 into the actual temperature measurement provided by sensor 108 . Similarly, system 200 may include voltage rounding logic 214 (e.g., to round current threshold and/or supply voltage values to the nearest value stored in storage unit 204) and voltage interpolation logic 218 (e.g., used to interpolate the value output by the storage unit 204 into the actual voltage value provided by the control logic 110 ). The multiplier 208 can take the determined temperature calibration factor (for example, from the storage unit 202 according to the detected temperature of the sensor 108), the determined voltage calibration factor (for example, according to the current voltage value provided by the logic 110, from the storage unit 204 lookup) and the reference leak value (from storage unit 206). This product value may then be used to manage the power set point (eg, via power management logic 112 ) as discussed in FIG. 1 .

Referring now to Figure 2B. System 250 may include a reference leakage storage unit 252 that stores reference leakage values for a corresponding set of voltages. Accordingly, in one embodiment, a single memory cell (252) may store multiple values that are comparable to the value stored in reference leakage memory cell 206 of FIG. 2A and the value stored in voltage scaling factor memory cell 204 of FIG. 2A Corresponds to the combination of corresponding values of . For example, the plurality of leakage power values may be indexed by temperature factors (eg, provided by sensor 108 ) and voltage factors (eg, corresponding to threshold voltage values and/or supply voltage values provided by logic 110 ). Such an embodiment may be provided by a single lookup (eg, based on the current threshold and/or supply voltage value from logic 110 ) Detected temperature value to find) the temperature calibration factor to calibrate the reference leak value. Alternatively, the values stored in storage units 202, 204, 206, and/or 252 may be combined into a single storage unit (not shown) to allow a single lookup to provide the sensed temperature value provided by sensor 108 And/or from the logic 110 the leakage value corresponding to the current threshold and/or supply voltage value. Also, system 250 may include rounding and/or interpolation logic (eg, may be the same as or similar to logic 210, 212, 214, and/or 218) according to some embodiments.

FIG. 3 shows a block diagram of a processor core 300 according to an embodiment. In one embodiment, core 300 may represent various components that may be present in a processor or several processors (such as those discussed with reference to FIGS. 5 and 6 ). Processor core 300 may include one or more domains, such as a second-level cache domain 302 , a front-end domain 304 , and one or more back-end domains 306 . Components within each of domains 302, 304, and 306 may be clocked by different clock signals as discussed with reference to FIG. 1 . Also, in different embodiments, each domain (eg, 302 , 304 , and 306 ) may include more or fewer components than those shown in FIG. 3 .

Level 2 (L2) cache domain 302 may include L2 cache 308 (eg, operable to store data including instructions), sensor 108 , and logic 106 , 110 , and 112 . In one embodiment, L2 cache 308 may be shared by multiple cores in the multi-core processor discussed with reference to FIGS. 5 and 6 . Also, L2 cache 308 may be from the same die as these processor cores. Therefore, in various embodiments of the present invention, the processor may include domains 304 and 306, and the processor may or may not include L2 cache 308.

As shown in FIG. 3, front-end domain 304 may include sensors 108, logic 106, 110, and 112, reorder buffer 318, rename and redirect unit 320, instruction cache 322, decode unit 324, sequencer 326, and/or Or one or more of branch prediction units 328 . In one embodiment, the front-end domain 304 may include other components, such as an instruction fetch unit.

The backend domain 306 may include a level one (L1) cache domain 328 and one or more of one or more execution domains 330 - 1 through 330 -N. L1 cache domain 328 may include L1 cache 332 (eg, to store data including instructions), sensor 108 , and logic 106 , 110 , and 112 . Furthermore, execution domains 330-1 through 330-N may include one or more integer execution units and/or floating point execution units. Each of execution domains 330-1 through 330-N may include an initiation queue (338-1 through 338-N, respectively), a register file (340-1 through 340-N, respectively), sensors 108, logic 106, 110 and 112, and/or execution units (346-1 to 346-N, respectively).

In one embodiment, each of domains 302, 304, and 306 may include one or more first-in-first-out (FIFO) buffers 348 to synchronize clock domains (eg, between domains 302, 304, and/or 306) between) communications.

Additionally, processor core 300 (and, in an embodiment (such as the embodiment shown in FIG. 3 ), backend domain 306 ) may include an interconnect or bus 350 to facilitate communication between the various components of processor core 300 . For example, after an instruction is successfully executed (eg, by execution domains 330 - 1 through 330 -N), the commit of the instruction may be communicated to ROB 318 (eg, via interconnect 350 ) to retire the instruction. Additionally, domains within the backend domain (eg, domain 328 and 330 - 1 through 330 -N) can communicate via interconnect 350 . For example, for type conversion instructions, communication between execution units ( 330 - 1 to 330 -N) may occur. Further operation of the components of FIGS. 1-3 will be discussed with reference to method 400 of FIG. 4 .

Additionally, while FIG. 3 shows that each of domains 302 , 304 , and 306 may include sensors 108 and logic 106 , 110 , and 112 , the domains may also share the same sensors 108 and logic 106 , 110 , and 112 . For example, all domains of processor core 300 may use only one set of sensors 108 and logic 106 , 110 , and 112 .

FIG. 4 shows a flowchart of a method 400 of providing an estimated leakage power, according to an embodiment. In one embodiment, the operations of method 400 may be performed by one or more components, such as those discussed with reference to FIGS. 1-3 and FIGS. 5 and 6 .

Now look at Figures 1 to 4. At operation 402, the sensor 108 may detect one or more temperature values corresponding to the IC device. The measured temperature value may be used at operation 404 to determine a temperature scaling factor (eg, determined using storage unit 202 ). At operation 404 , a voltage scaling factor may also be determined (eg, determined using storage units 204 and/or 252 ) as discussed with reference to FIGS. 2A and 2B . At operation 406 , the scaling factor determined at operation 404 may be used to scale the reference leakage value (eg, stored by units 206 and/or 252 ) as discussed with reference to FIGS. 2A and 2B . At operation 408 , a signal corresponding to the estimated leakage power of the IC device may be generated (eg, via multipliers 205 and 254 ). As discussed with reference to FIG. 1 , the estimated leakage power ( 408 ) may be used to adjust power consumption of one or more components of a computing system (such as the systems discussed with reference to FIGS. 1 , 5 , and/or 6 ). .

In an embodiment, the following equation is used to provide the estimated leakage power at operation 408:

$\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\frac{{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}}}{{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; +</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; \&delta;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>}$

相应的值可以被存储到存储单元202，而和In the above formula, P corresponds to the estimated leakage power value, P ₀ corresponds to the reference leakage power value (such as may be stored in units 206 and/or 252), V _dd corresponds to the supply voltage (which may be provided by logic 110 ), V _th corresponds to the threshold voltage (which may be provided by logic 110 ), V _dd0 corresponds to V _dd when measuring reference leakage, V _tho corresponds to V _th when measuring reference leakage, and T corresponds to To the current temperature value, T ₀ corresponds to the temperature at which the reference leak is measured, and δ, β, and γ are all circuit-dependent constants set by the designer. In various embodiments, and

The corresponding value can be stored to storage unit 202, and the

$\mathrm{<span\; class="notranslate">\; V</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}}}{{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; dd</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; +</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; )</span>}$

The corresponding value may be stored to storage unit 204 (or 252). Therefore, multipliers ( 208 , 254 ) can be used to multiply T(T) and V(V _dd , V _th ) to provide a value for P.

Furthermore, in one embodiment, dynamic calibration of IC components may also be performed in idle mode (eg, when there is no dynamic power consumption). In this case, the temperature rise (relative to the controlled ambient temperature) within each portion of the IC component (eg, functional blocks) may be related to the leakage power. Thermal sensors 108, which can be installed among the functional blocks, can report this stable temperature (eg, after a relatively long time interval). Using this temperature map, a tool (eg, a computing device external to the IC component) can derive the power map that caused this situation, eg via reverse engineering. Leakage values can then be calculated from the quiescent temperatures of the parts (since other constants may be known, such as supply voltage, threshold voltage, and ambient temperature). Once the power map is calculated, it can be stored in the reference leakage storage unit 206. In an embodiment, special dedicated microcode may be used to communicate between the IC component being calibrated and the test equipment (for example, to report these temperature readings and perform baseline leakage updates).

FIG. 5 shows a block diagram of a computing system 500 according to an embodiment of the present invention. Computing system 500 may include one or more central processing units (CPUs) 502 or processors in communication via an interconnection network (or bus) 504 . Processor 502 may be any type of processor, such as a general-purpose processor, a network processor (a processor that processes data communicated over computer network 503), or other types of processors (including Reduced Instruction Set Computer (RISC ) processor or Complex Instruction Set Computer (CISC) processor). Also, processor 502 may have a single-core or multi-core design. Processor 502 with a multi-core design may integrate different types of processor cores on the same integrated circuit (IC) die. Furthermore, processor 502 having a multi-core design may also be implemented as a symmetric or asymmetric multiprocessor. In an embodiment, one or more processors 502 may utilize the embodiments discussed with reference to FIGS. 1-4 . For example, one or more processors 502 may include one or more processor cores (300). Moreover, the operations discussed with reference to FIGS. 1-4 may also be performed by one or more components of the system 500 .

Chipset 506 is also capable of communicating with interconnection network 504 . Chipset 506 may include a memory hub (MCH) 508 . MCH 508 may include memory controller 510 in communication with memory 512 . Memory 512 may store data as well as sequences of instructions executed by CPU 502 or any other device included in computing system 500 . In one embodiment of the present invention, the memory 512 may include one or more volatile storage devices (memory), such as random access memory (RAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), static RAM ( SRAM), etc., can also use non-volatile memory, such as hard disk. Other devices may also communicate over the interconnection network 504, such as multiple CPUs and/or multiple system memories.

The MCH 508 also includes a graphics interface 514 for communicating with a graphics accelerator 516 . In one embodiment of the invention, graphics interface 514 may communicate with graphics accelerator 516 via an accelerated graphics port (AGP). In an embodiment of the invention, a display (such as a flat panel display) may communicate via, for example, a signal converter and graphics interface 512 that converts a digital representation of an image stored in a storage device (such as video memory or system memory) to method into a display signal that can be interpreted and displayed by the display. The display signals produced by the display device need to pass through various control devices before being interpreted and subsequently displayed on the display.

Central interface 518 enables MCH 508 to communicate with input/output control center (ICH) 520 . ICH 520 may provide an interface for I/O devices that communicate with components of computing system 500 . ICH 520 may communicate via peripheral bridge (or controller) 524 and bus 522 , such as via a peripheral component interconnect (PCI) bridge, a universal serial bus (USB) controller, or the like. Bridge 524 may provide a data path between CPU 502 and peripheral devices. Other types of topologies are also possible. Also, multiple buses can communicate with the ICH520, for example via multiple bridges or controllers. Additionally, in various embodiments of the present invention, other peripherals that communicate with the ICH520 include Integrated Drive Electronics (IDE) or Small Computer System Interface (SCSI) hard drives, USB ports, keyboards, mice, parallel ports, serial ports, floppy drives, digital output support devices such as Digital Visual Interface (DVI), and so on.

Bus 522 may communicate with audio device 526, one or more disk drives 528, and network interface device 530 (which in turn communicates with computer network 503). Other devices may communicate with bus 522 . Moreover, in some embodiments of the present invention, various components (such as the network interface device 530 ) may communicate with the MCH 508 . Alternatively, processor 502 and MCH 508 may be combined to form a single chip. Furthermore, in other embodiments of the present invention, graphics accelerator 516 may be included in MCH 508 .

Additionally, computing system 500 may include volatile and/or nonvolatile memory (or storage). Nonvolatile memory can include, for example, one or more of the following devices: Read Only Memory (ROM), Programmable ROM (PROM), Erasable PROM (EPROM), Electronic EPROM (EEPROM), Diskette Drive, floppy disk, compact disk ROM (CD-ROM), digital versatile disk (DVD), flash memory, magneto-optical disk, or other type of non-volatile machine-readable medium capable of storing electronic instructions and/or data.

FIG. 6 illustrates a computing system 600 designed in a point-to-point (PtP) architecture, according to an embodiment of the present invention. In particular, Figure 6 shows a system in which processors, memory, and input/output devices are interconnected by several point-to-point interfaces. The operations discussed with reference to FIGS. 1-5 may be performed by one or more components of system 600 .

As shown in FIG. 6, system 600 may include several processors, although only two processors are shown: processors 602 and 604, for clarity. Processors 602 and 604 may each include a local memory hub (MCH) 606 and 608 to communicate with memories 610 and 612 . Memory 610 and/or 612 may store various data, such as those discussed with reference to memory 512 .

Processors 602 and 604 may be any type of processor, such as the type discussed with reference to processor 502 of FIG. 5 . Processors 602 and 604 may employ PtP interface circuits 616 and 618 via a point-to-point (PtP) interface 614 to exchange data. Processors 602 and 604 may each employ point-to-point interface circuits 626 , 628 , 630 and 632 and chipset 620 to exchange data via separate PtP interfaces 622 and 624 . Chipset 620 may also employ PtP interface circuit 637 and high-performance graphics circuit 634 to exchange data via high-performance graphics interface 636 .

At least one embodiment of the invention may be implemented within processors 602 and 604 . For example, one or more of domains 102 and/or processor core 300 discussed with reference to FIG. 1 may be located within processors 602 and 604 . However, other circuits, logic units, or devices within the system 600 of FIG. 6 may also employ other embodiments of the present invention. In addition, other embodiments of the present invention may be distributed among several circuits, logic units or devices shown in FIG. 6 .

The chipset 620 can use a PtP interface circuit 641 to communicate with the bus 640 , and the bus 640 can have one or more devices communicating with it, such as a bus bridge 642 and an I/O device 643 . The bus bridge 643 can communicate with other devices via the bus 644, such as a keyboard/mouse 645, a communication device 646 (such as a modem that can communicate with the computer network 503, a network interface device, etc.), audio I/O devices, and/or data Storage device 648. Data storage 648 may store code 649 for execution by processor 602 and/or 604 .

In various embodiments of the present invention, the operations discussed herein (such as those discussed with reference to FIGS. 1 to 6 ) may be implemented by software, firmware, microcode, hardware (circuits), or a combination thereof as a computer program product. For example, the products include machine-readable or computer-readable media having stored thereon instructions for programming a computer to perform the processes described herein. Also, by way of example, the term "logic" may include software, hardware, firmware, or a combination of software and hardware. The machine-readable medium may include a storage device such as those discussed with reference to FIGS. 1-6 . In addition, these computer-readable media can be downloaded as a computer program product, wherein the program can be transmitted from a remote computer ( such as a server) to a requesting computer (such as a client). Therefore, herein, carrier waves will be considered to include machine-readable media.

In this specification, reference to "one embodiment" or "an embodiment" means that the characteristic, structure or characteristic described with respect to the embodiment can be included in at least the implementation. The appearances of the phrase "in one embodiment" in various places in this specification may or may not all refer to the same embodiment.

Also, in the description and claims, the terms "coupled" and "connected", along with their derivatives, may be used. In some embodiments of the invention, "connected" may be used to mean that two or more elements are in direct physical or electrical contact with each other. "Coupled" may mean that two or more elements are in direct physical or electrical contact; however, "coupled" may also mean that two or more elements may not be in direct contact with each other, but are still capable of coordinating or interacting with each other.

Therefore, although embodiments of the invention have been described in language specific to structural features and/or functional acts, it is to be understood that claimed subject matter may not be limited to the specific features or acts described. action. Rather, the specific features and acts are disclosed as example forms of implementing the claimed subject matter.

Claims (1)

1. A processor comprising: Multiple cores, where each core includes: an instruction fetch unit for fetching instructions; an instruction decoding unit for decoding said instruction; branch prediction unit; register file; A cache with multiple cache levels, including a first-level (L1) cache and a second-level (L2) cache; a plurality of execution units for executing said instructions; where each core is organized into multiple domains, each operating at a specified voltage and frequency, wherein the L2 cache is assigned to a first domain having a first voltage and frequency, and at least one other logic unit of the processor is assigned to a second domain having a second voltage and frequency; Logic to perform actions related to power usage while determining the impact of leakage.

Images