(19)
(11)EP 3 283 966 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
27.11.2019 Bulletin 2019/48

(21)Application number: 16719624.5

(22)Date of filing:  14.04.2016
(51)Int. Cl.: 
G06F 12/0862  (2016.01)
G06F 12/02  (2006.01)
G06F 9/455  (2018.01)
G06F 12/109  (2016.01)
(86)International application number:
PCT/US2016/027420
(87)International publication number:
WO 2016/168395 (20.10.2016 Gazette  2016/42)

(54)

VIRTUALIZATION-AWARE PREFETCHING

VIRTUALISIERUNGSBEWUSSTES VORAUSLADEN

PRÉ-EXTRACTION TENANT COMPTE DE LA VIRTUALISATION


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 14.04.2015 US 201514685673

(43)Date of publication of application:
21.02.2018 Bulletin 2018/08

(73)Proprietor: Google LLC
Mountain View, CA 94043 (US)

(72)Inventors:
  • YOO, Richard
    Mountain View, California 94043 (US)
  • CHENG, Liqun
    Mountain View, California 94043 (US)
  • RANGANATHAN, Parthasarathy
    Mountain View, California 94043 (US)
  • GOVINDARAJU, Rama Krishna
    Mountain View, California 94043 (US)

(74)Representative: Betten & Resch 
Patent- und Rechtsanwälte PartGmbB Maximiliansplatz 14
80333 München
80333 München (DE)


(56)References cited: : 
US-A1- 2003 105 937
US-A1- 2015 082 000
US-A1- 2015 039 838
  
      
    Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


    Description

    BACKGROUND



    [0001] This specification relates to pre-fetching content for a cache.

    [0002] Some processors include a hardware pre-fetcher that automatically determines data, e.g., instructions or other types of data that are likely to be used soon by the processor. A pre-fetcher may use various mechanisms to predict a likelihood of reuse or predict which data to pre-fetch. In some examples, the pre-fetcher may determine data stored adjacent to data currently being used, and which is not already stored in a cache, and pre-fetch the determined data, e.g., using a physical address of a device that includes the processor.

    [0003] US 2015/0039838 discloses a system that facilitates reduction of latency associated with accessing content of a memory page that has been swapped out by a guest operating system in a virtualized computer system. During operation, a hypervisor detects an I/O write command issued by the guest operating system at a swap location within the guest operating system's swap file and records the swap location. The hypervisor then prefetches contents of a page stored at the swap location within the guest operating system's swap file into a prefetch cache in host machine memory. Subsequently, the hypervisor detects an I/O read command issued by the guest operating system at the swap location within the swap file. In response, the hypervisor provides contents of the page to the guest operating system from the prefetch cache, thereby avoiding accessing the guest operating system's swap file.

    SUMMARY



    [0004] The invention is set out in the appended set of claims. The dependent claims set out particular embodiments. The embodiments or examples of the following description which are not covered by the appended claims are considered as not being part of the invention according to this description.

    [0005] In some implementations, a pre-fetcher uses a virtual machine's physical addresses (and not virtual machine's virtual addresses or the physical addresses of the actual hardware executing the virtual machine) to determine which pages, e.g., cache lines, to pre-fetch for a cache. The pre-fetcher may use a register value to determine which type of prefetching to perform, e.g., using a virtual machine's physical addresses or hardware physical addresses, depending on the application for which the pre-fetcher is getting data, e.g., when a single pre-fetcher fetches data for virtual machines and native applications. In some examples, a pre-fetcher may not have access to the virtual machine's physical addresses, so the pre-fetcher may need to be given access to L1 request traffic or other data that includes the virtual machine's physical addresses.

    [0006] In general, one innovative aspect of the subject matter described in this specification can be embodied in methods that include the actions of determining, for a virtual machine executing on a device and by a pre-fetcher using a first virtual machine physical address associated with the virtual machine, a second virtual machine physical address for data to pre-fetch for the execution of the virtual machine on the device, determining, using the second virtual machine physical address and an address mapping that associates virtual machine physical addresses for the virtual machine with device physical addresses for the device, a device physical address for the data, and requesting the data from a memory using the device physical address. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

    [0007] A system of one or more computers can be configured to perform particular operations or actions by virtue of having software, firmware, hardware, or a combination of them installed on the system that in operation causes or cause the system to perform the actions. One or more computer programs can be configured to perform particular operations or actions by virtue of including instructions that, when executed by data processing apparatus, cause the apparatus to perform the actions.

    [0008] In general, one innovative aspect of the subject matter described in this specification can be embodied in a system that includes a pre-fetcher configured to perform operations including determining, for a virtual machine executing on a device and using a first virtual machine physical address associated with the virtual machine, a second virtual machine physical address for data to pre-fetch for the execution of the virtual machine on the device, determining, using the second virtual machine physical address and an address mapping that associates virtual machine physical addresses for the virtual machine with device physical addresses for the device, a device physical address for the data, and requesting the data from a memory using the device physical address. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

    [0009] The foregoing and other embodiments can each optionally include one or more of the following features, alone or in combination. In some implementations, requesting the data from the memory using the device physical address includes requesting the data from a memory that is not a cache of the device using the device physical address. The method may include receiving, by the pre-fetcher, the first virtual machine physical address. Receiving, by the pre-fetcher, the first virtual machine physical address may include using, by the pre-fetcher, a cache port to retrieve the first virtual machine physical address, wherein a cache includes the cache port used by the pre-fetcher and another cache port used by a processor of the device. Receiving, by the pre-fetcher, the first virtual machine physical address may include using, by the pre-fetcher, a cache port to retrieve the first virtual machine physical address, wherein the cache port is used by the pre-fetcher and by a processor of the device to access a cache. Receiving, by the pre-fetcher, the virtual machine physical address may include receiving the virtual machine physical address from a processor of the device.

    [0010] In some implementations, the method may include receiving the data from the memory in response to requesting the data from the memory using the device physical address, and storing the data in a cache. The method may include receiving the data from the memory in response to requesting the data from the memory using the device physical address, and storing the data in a buffer. The method may include determining, for each of multiple applications executing on the device, whether the application is a virtual machine, in response to determining for each application of the multiple applications that is a virtual machine, determining the second virtual machine physical address for the data to pre-fetch, and, in response to determining for each application of the multiple applications that is not a virtual machine, determining, by the pre-fetcher using a second device physical address, a third device physical address for other data to pre-fetch. Determining, for each of the multiple applications executing on the device, whether the application is a virtual machine may include determining whether a register on a processor of the device indicates that application is a virtual machine. The method may include detecting, for each application of the multiple applications that is a virtual machine, the execution of a launch instruction or a resume instruction to execute the virtual machine on the device, and in response to detecting the execution of the launch instruction or the resume instruction to execute the virtual machine on the device, setting the register to a value that indicates that the processor of the device is currently executing the virtual machine. The method may include detecting, for each application of the multiple applications that is a virtual machine, the execution of an exit instruction during the execution of the virtual machine on the device, and in response to detecting the execution of the exit instruction during the execution of the virtual machine on the device, setting the register to another value that indicates that the processor of the device is not currently executing the virtual machine.

    [0011] In some implementations, determining the second virtual machine physical address for the data to pre-fetch for the execution of the virtual machine on the device may include determining the second virtual machine physical address for an instruction to pre-fetch for the execution of the virtual machine on the device. Requesting the data from the memory using the device physical address may include requesting the instruction from the memory using the device physical address.

    [0012] In some implementations, requesting the data from the memory using the device physical address may include requesting the data from a memory that is not a cache of the device using the device physical address. The operations may include receiving the first virtual machine physical address. The system may include a processor on the device, and a cache on the processor including i) a first cache port configured to allow the pre-fetcher access to virtual machine physical addresses, ii) and a second cache port configured to allow the processor access to the cache, wherein receiving, by the pre-fetcher, the first virtual machine physical address may include using, by the pre-fetcher, the first cache port to retrieve the first virtual machine physical address.

    [0013] In some implementations, the system may include a processor on the device, and a cache on the processor including a cache port configured to allow the pre-fetcher access to virtual machine physical addresses on the cache and allow the processor access to the cache, wherein receiving, by the pre-fetcher, the first virtual machine physical address may include using, by the pre-fetcher, the cache port to retrieve the first virtual machine physical address. The system may include a processor on the device configured to provide the virtual machine physical address to the pre-fetcher, wherein receiving, by the pre-fetcher, the virtual machine physical address may include receiving the virtual machine physical address from the processor.

    [0014] In some implementations, the system may include a cache to store the requested data received from the memory. The system may include a buffer to store the requested data received from the memory. The system may include a second memory to store the address mapping that associates virtual machine physical addresses for the virtual machine with device physical addresses for a device. The system may include a processor that includes the second memory, the cache, and the pre-fetcher.

    [0015] In some implementations, the operations may include determining, for each of multiple applications executing on the device, whether the application is a virtual machine, in response to determining for each application of the multiple applications that is a virtual machine, determining the second virtual machine physical address for the data to pre-fetch, and in response to determining for each application of the multiple applications that is not a virtual machine, determining, using a second device physical address, a third device physical address for other data to pre-fetch.

    [0016] In some implementations, the system may include a register configured to store a value that indicates whether or not an application executing on a corresponding processor of the device is a virtual machine, wherein determining, for each of the multiple applications executing on the device, whether the application is a virtual machine may include determining whether the register on a processor of the device indicates that application is a virtual machine. The system may include a processor configured to: detect, for each application of the multiple applications that is a virtual machine, the execution of a launch instruction or a resume instruction to execute the virtual machine on the device, and in response to detecting the execution of the launch instruction or the resume instruction to execute the virtual machine on the device, set the register to a value that indicates that the processor of the device is currently executing the virtual machine. The processor may be configured to: detect, for each application of the multiple applications that is a virtual machine, the execution of an exit instruction during the execution of the virtual machine on the device, and in response to detecting the execution of the exit instruction during the execution of the virtual machine on the device, set the register to another value that indicates that the processor of the device is not currently executing the virtual machine.

    [0017] In some implementations, determining the second virtual machine physical address for the data to pre-fetch for the execution of the virtual machine on the device may include determining the second virtual machine physical address for an instruction to pre-fetch for the execution of the virtual machine on the device, and requesting the data from the memory using the device physical address may include requesting the instruction from the memory using the device physical address.

    [0018] The subject matter described in this specification can be implemented in particular embodiments and may result in one or more of the following advantages. In some implementations, a pre-fetcher that uses virtual machine physical addresses to pre-fetch data may pre-fetch multiple consecutive pages of data using virtual machine physical addresses, apply different pre-fetching algorithms to different virtual machines, or both. In some implementations, the pre-fetcher may have a higher effectiveness than other types of pre-fetchers that do not use virtual machine physical addresses to pre-fetch data, may reduce power consumption over the other types of pre-fetches, or both. In some implementations, when different pre-fetching algorithms are applied to different virtual machines, the characteristics of each virtual machine, its workload, and/or its memory accesses may be taken into account individually by the pre-fetcher so that prefetching may be more effective for each virtual machine. In some implementations, interference of prefetching for a virtual machine with other data requests may be reduced, when the other data requests may be due to prefetching for another virtual machine or demand requests.

    [0019] The details of one or more embodiments of the subject matter of this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.

    BRIEF DESCRIPTION OF THE DRAWINGS



    [0020] 

    FIG. 1 is an example of a system in which a pre-fetcher pre-fetches data using a virtual machine physical address.

    FIG. 2 shows an example of address translation in a virtualized environment.

    FIG. 3 is a flow diagram of a process for storing pre-fetched data in a memory.

    FIG. 4 is a block diagram of a computing system that can be used in connection with computer-implemented methods described in this document.



    [0021] Like reference numbers and designations in the various drawings indicate like elements.

    DETAILED DESCRIPTION



    [0022] A pre-fetching system may allow a pre-fetcher to change its behavior depending on whether an application a processor is currently executing is a virtual machine. For instance, when a processor begins executing a virtual machine, e.g., using VMLAUNCH or VMRESUME, the processor may set a value of a register to indicate that the processor is executing a virtual machine. When the processor stops executing the virtual machine, e.g., based on VMEXIT, the processor may set the value of the register to indicate that the processor is not executing the virtual machine, and may change the value of the register as necessary based on a next application the processor executes.

    [0023] When the register is not set, e.g., and indicates that a virtual machine is not being executed, the pre-fetcher makes pre-fetch predictions based on a host physical address stream for a host that includes the processor, e.g., a computer. When the register is set, e.g., and indicates that a virtual machine is being executed, the pre-fetcher makes pre-fetch predictions using a guest physical address stream, e.g., of the virtual machine.

    [0024] When the registers is set, the pre-fetcher determines a candidate guest physical address for data to pre-fetch and uses a mapping between guest physical addresses and host physical addresses to translate the candidate guest physical address to a host physical address. In some examples, the pre-fetcher may have access to a page walker that provides the pre-fetcher with the guest physical address to host physical address mapping, e.g., using a nested page walk routine. In some implementations, e.g., when a processor maintains separate caches for nested page table lookup, a pre-fetcher may have access, e.g., using a port, to the separate cache for nested page table lookup.

    [0025] Once the pre-fetcher determines the host physical address, the pre-fetcher sends the host physical address down the memory hierarchy to perform pre-fetch. In some examples, the pre-fetcher may include a pre-fetch priority tag in a pre-fetch request with the host physical address. Data identified by the host physical address may then be stored in a cache or a buffer.

    [0026] FIG. 1 is an example of a system 100 in which a pre-fetcher pre-fetches data using a virtual machine physical address. For instance, the system 100 includes an operating system 102 that may execute a virtual machine 104, an application 110, both, or a combination of multiple virtual machines and multiple applications.

    [0027] Each virtual machine includes multiple virtual machine virtual addresses 106 and multiple virtual machine physical addresses 108, the latter of which may be mapped to device physical addresses using an address mapping 118. FIG. 2 shows an example of address translation in a virtualized environment 200, and may be used by the address mapping 118. For instance, a virtual machine 202 maps virtual machine virtual addresses 204 to virtual machine physical address 206. When the virtual machine 202 or an application executing on the virtual machine requires data, the virtual machine 202 provides a system, e.g., a memory management unit, with a virtual machine virtual address 204. The system translates the virtual machine virtual address 204 to a virtual machine physical address 206 and then translates the virtual machine physical address 206 to a host physical address 208, e.g., using nested page tables. The system may then request data identified using the host physical address 208 and provide the data to the virtual machine 202.

    [0028] For example, referring to FIG. 1, the virtual machine 104, at time TA, requests data from a processor 112. The request may include a virtual machine virtual address 106 or a virtual machine physical address 108. The requested data may include information bits on which a sequence of operators, e.g., "instructions," operate. In some examples, the requested data may include instructions, e.g., the request may be a request for instructions.

    [0029] A processor 112 acts on the request, e.g., identifies data responsive to the request. For instance, the processor 112 may check a cache 116 to determine whether the cache 116 includes the requested data. In some examples, the processor 112 may confirm that a virtual machine register 114 is set when acting on the request. In some examples, the processor 112 sets the virtual machine register 114 when the processor 112 begins execution of a virtual machine, irrespective of the virtual machine being executed, and unsets the virtual machine register 114 when the processor 112 stops executing the virtual machine, e.g., in response to a pause or exit command.

    [0030] The processor 112, or a component of the processor, e.g., a memory management unit, may use an address mapping 118 to determine a device physical address for the requested data. For instance, the processor 112 may use nested page tables to determine the device physical address using a virtual machine virtual address 106. In some examples, the processor 112 may use a page table to determine the device physical address using a virtual machine physical address 108. The address mapping 118 may be included on the processor 112 or another component of the system 100.

    [0031] The processor 112 determines whether the requested data is in the cache 116 or should be requested from another memory. The processor 112 may retrieve the requested data and provide the requested data to the virtual machine 104.

    [0032] At time TB, a pre-fetcher 120 determines whether the processor 112 is acting on a request from a virtual machine. For instance, the pre-fetcher 120 may check the virtual machine register 114 to determine whether a bit is set that indicates that the processor 112 is acting on a request from a virtual machine. The pre-fetcher 120 may use any appropriate method to determine whether the processor 112 is acting on a request from a virtual machine.

    [0033] The pre-fetcher 120 may determine whether the processor 112 is acting on a request from a virtual machine any time after the processor 112 receives the data request. For instance, the processor 112 maybe retrieving the requested data or performing an operation on or with the requested data while the pre-fetcher 120 determines whether the processor is acting on a request from a virtual machine.

    [0034] When the pre-fetcher 120 determines that the processor 112 is acting on a request from a virtual machine, whether the virtual machine 104 or another virtual machine executing in the system 100, the pre-fetcher 120, during time Tc, determines the virtual machine physical address that corresponds to the data request. The pre-fetcher 120 may access a virtual address stream in the processor 112, such as L1 cache request traffic, to determine the virtual machine physical address, e.g., by receiving a virtual machine virtual address and using the address mapping to determine the virtual machine physical address. In some implementations, the pre-fetcher 120 determines the virtual machine physical address or the virtual machine virtual address from the processor 112. The pre-fetcher 120 may use any appropriate method to determine the virtual machine physical address.

    [0035] In some examples, the pre-fetcher 120 may have access to the virtual machine virtual addresses included in requests sent to the processor 112 and not the virtual machine physical addresses. In these examples, the pre-fetcher 120 may use the address mapping 118 to determine a virtual machine physical address using a virtual machine virtual address included in the data request.

    [0036] Once the pre-fetcher 120 determines the virtual machine physical address for the data request, the pre-fetcher 120 determines a candidate virtual machine physical address, at time TD, for additional data that should be pre-fetched and stored in the cache 116 or a buffer, e.g., a next cache line to pre-fetch. For instance, the pre-fetcher 120 may determine a candidate address adjacent to the virtual machine physical address and use the candidate address to pre-fetch the additional data. In some examples, the pre-fetcher 120 may use an algorithm that predicts, using the virtual machine physical address, which additional data may be requested by the virtual machine 104 and uses the virtual machine physical address for that additional data as the candidate address. In some implementations, the pre-fetcher 120, e.g., a spatial pre-fetcher, can use a bit vector representing which data should be pre-fetched. The pre-fetcher may use the bit vector, combined with a base address requested by a trigger event, e.g., the virtual machine physical address included in the data request at time TA, to determine a list of addresses to pre-fetch. The pre-fetcher 120 may use any appropriate algorithm to determine the candidate address, e.g., to perform pre-fetching using a virtual machine physical address. The pre-fetcher 120 may determine the candidate address whether or not the processor 112 retrieves the requested data from the cache 116.

    [0037] At time TE, the pre-fetcher 120 determines a device physical address for the candidate virtual machine physical address using the address mapping 118. The pre-fetcher 120, at time TF, uses the device physical address to pre-fetch the additional data. For instance, the pre-fetcher 120 sends a request for the additional data to a memory 122 associated with the device physical address or otherwise causes the pre-fetching of the additional data from the memory 122. In some examples, the memory 122 is not a cache in the system 100 but may include a cache. For instance, the memory may be a random access memory, or any other appropriate type of memory.

    [0038] The memory 122 uses the device physical address to retrieve the additional data. The memory 122, or another component in the system 100, stores the additional data, at time TG, in the cache 116 or a buffer. In some examples, the memory 122 provides the additional data to the processor 112 and a component of the processor 112 stores the additional data in the cache 116 or the buffer. The buffer may be a storage for pre-fetched data.

    [0039] When the pre-fetcher 120 determines that the processor 112 is not acting on a request from a virtual machine and is acting on a request from an application, whether the application 110 or another application executing in the system 100, the pre-fetcher 120 uses a device physical address that corresponds to the requested data to determine a candidate device physical address for other additional data to pre-fetch for the application. The pre-fetcher 120 may use any appropriate algorithm to determine the candidate device physical address using the device physical address, e.g., to perform pre-fetching for a native application of the operating system 102. In some examples, the pre-fetcher 120 may use information about relative priorities of various jobs when making a decision of what data should be pre-fetched. For instance, the pre-fetcher 120 may use priority information for different applications executing on the system 100 to determine for which application data should be pre-fetched.

    [0040] The pre-fetcher 120 may send the candidate device physical address to a memory, e.g., the memory 122, to pre-fetch the other additional data. The system 100, e.g., the processor 112 or the memory 122, may store the other additional data in the cache 116 or a buffer.

    [0041] In some implementations, the pre-fetcher 120 is part of the processor 112, e.g., the pre-fetcher 120 is a hardware pre-fetcher. In some implementations, the pre-fetcher 120 is a software pre-fetcher.

    [0042] FIG. 3 is a flow diagram of a process 300 for storing pre-fetched data in a memory. For example, the process 300 can be used by the pre-fetcher 120 from the system 100.

    [0043] A pre-fetcher determines whether an application is a virtual machine (302). For instance, the pre-fetcher checks a value, e.g., a bit, on a processor to determine whether the application is a virtual machine. In some implementations, the pre-fetcher resides on the processor. In some examples, the pre-fetcher is located on another portion of a device that includes the processor.

    [0044] In response to determining that the application is a virtual machine, the pre-fetcher determines a first virtual machine physical address for data requested by the virtual machine (304). For instance, the pre-fetcher accesses a virtual address stream of the virtual machine. The pre-fetcher determines a virtual machine virtual address from the virtual address stream and uses the virtual address to determine a virtual machine physical address.

    [0045] In some examples, the pre-fetcher has access to a cache port of a cache on the processor executing the virtual machine. The pre-fetcher accesses the cache port to determine the virtual machine virtual address or the virtual machine physical address. In some implementations, the pre-fetcher uses the same cache port as the processor to access the cache. In these implementations, a system may use relative priority schemes when allowing the pre-fetcher and the processor access to the cache using the same cache port. In some implementations, the pre-fetcher uses a different cache port than a cache port used by the processor to access the cache.

    [0046] The pre-fetcher determines, using the first virtual machine physical address, a second virtual machine physical address for data to store in a cache (306). The pre-fetcher may use any appropriate algorithm to determine the second virtual machine physical address for pre-fetching data.

    [0047] The pre-fetcher determines, using a mapping of virtual machine physical addresses to device physical addresses, a device physical address (308). For instance, the pre-fetcher uses an address mapping, such as a page table or a nested page table, to determine the device physical address.

    [0048] The pre-fetcher requests the data from a memory using the device physical address (310). For example, the pre-fetcher requests the data from a random access memory.

    [0049] The pre-fetcher stores the data in another memory (312). The other memory may include a cache on the processor, a buffer, e.g., on the processor, or another memory that has a lower latency than the memory from which the data was pre-fetched.

    [0050] In response to determining that the application is not a virtual machine, the pre-fetcher determines a first physical address for data requested by the application (314). For instance, the application may be an office productivity application, a web browser, or a game, and the pre-fetcher determines the first physical address for that application.

    [0051] The pre-fetcher determines, using the first physical address, a second physical address for data to pre-fetch (316). The pre-fetcher may use any appropriate algorithm to determine the second physical address for pre-fetching data.

    [0052] The pre-fetcher requests the data from the memory using the second device physical address (318). The pre-fetcher then stores the data in the other memory, e.g., a cache or a buffer. For instance, the pre-fetcher performs step 312.

    [0053] In some implementations, the process 300 can include additional steps, fewer steps, or some of the steps can be divided into multiple steps. For example, the pre-fetcher may perform steps 302 through 312 and not steps 314 through 318. In some examples, the pre-fetcher may perform steps 304 through 308 or steps 304 through 310 without performing the other steps.

    [0054] In some implementations, a pre-fetching system may include other appropriate mechanisms to allow the pre-fetcher to determine that a processor is executing a virtual machine. For instance, the pre-fetcher may read a value of a register to determine that the processor is executing a virtual machine.

    [0055] In some implementations, a pre-fetching system may include multiple levels of address layers, e.g., virtual machines executing within virtual machines, and a pre-fetcher may determine what data to pre-fetch using information about the multiple levels of address layers. For instance, the pre-fetcher may determine a highest address layer, e.g., for the virtual machine executing in the level furthest from the operating system executing on the pre-fetching system, and use information about physical addresses for that highest address layer when making pre-fetching decisions.

    [0056] In some implementations, when a single system, e.g., device or processor, executes multiple virtual machines, a pre-fetcher may apply different pre-fetching algorithms to each of the virtual machines or to some of the virtual machines. A pre-fetching system may include appropriate mechanisms to allow the pre-fetcher to determine which of the multiple virtual machines is executing to apply the corresponding pre-fetching algorithm. For instance, the pre-fetcher may pre-fetch data for every data request for a first virtual machine and pre-fetch data for every tenth data request for a second virtual machine. In some examples, the pre-fetcher may delay or omit pre-fetch requests or both to control bandwidth, e.g., the number of pre-fetch requests sent, or quality of service guarantees, or both.

    [0057] Embodiments of the subject matter and the functional operations described in this specification can be implemented in digital electronic circuitry, in tangibly-embodied computer software or firmware, in computer hardware, including the structures disclosed in this specification. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible non-transitory program carrier for execution by, or to control the operation of, data processing apparatus. Alternatively or in addition, the program instructions can be encoded on an artificially-generated propagated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal, that is generated to encode information for transmission to suitable receiver apparatus for execution by a data processing apparatus. The computer storage medium can be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or a combination of one or more of them.

    [0058] The term "data processing apparatus" refers to data processing hardware and encompasses all kinds of apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. The apparatus can also be or further include special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application-specific integrated circuit). The apparatus can optionally include, in addition to hardware, code that creates an execution environment for computer programs, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of one or more of them.

    [0059] A computer program, which may also be referred to or described as a program, software, a software application, a module, a software module, a script, or code, can be written in any form of programming language, including compiled or interpreted languages, or declarative or procedural languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data, e.g., one or more scripts stored in a markup language document, in a single file dedicated to the program in question, or in multiple coordinated files, e.g., files that store one or more modules, sub-programs, or portions of code. A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

    [0060] The processes and logic flows described in this specification can be performed by one or more programmable computers executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application-specific integrated circuit).

    [0061] Computers suitable for the execution of a computer program include, by way of example, general or special purpose microprocessors or both, or any other kind of central processing unit. Generally, a central processing unit will receive instructions and data from a read-only memory or a random access memory or both. The essential elements of a computer are a central processing unit for performing or executing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto-optical disks, or optical disks. However, a computer need not have such devices. Moreover, a computer can be embedded in another device, e.g., a mobile telephone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a Global Positioning System (GPS) receiver, or a portable storage device, e.g., a universal serial bus (USB) flash drive, to name just a few.

    [0062] Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

    [0063] An example of one such type of computer is shown in FIG. 4, which shows a schematic diagram of a generic computer system 400. The system 400 can be used for the operations described in association with any of the computer-implement methods described previously, according to one implementation. The system 400 includes a processor 410, a memory 420, a storage device 430, and an input/output device 440. Each of the components 410, 420, 430, and 440 are interconnected using a system bus 450. The processor 410 is capable of processing instructions for execution within the system 400. In one implementation, the processor 410 is a single-threaded processor. In another implementation, the processor 410 is a multi-threaded processor. The processor 410 is capable of processing instructions stored in the memory 420 or on the storage device 430 to display graphical information for a user interface on the input/output device 440.

    [0064] The memory 420 stores information within the system 400. In one implementation, the memory 420 is a computer-readable medium. In one implementation, the memory 420 is a volatile memory unit. In another implementation, the memory 420 is a non-volatile memory unit.

    [0065] The storage device 430 is capable of providing mass storage for the system 400. In one implementation, the storage device 430 is a computer-readable medium. In various different implementations, the storage device 430 may be a floppy disk device, a hard disk device, an optical disk device, or a tape device.

    [0066] The input/output device 440 provides input/output operations for the system 400. In one implementation, the input/output device 440 includes a keyboard and/or pointing device. In another implementation, the input/output device 440 includes a display unit for displaying graphical user interfaces.


    Claims

    1. A method comprising:

    determining (302), for a first application (110) executing on a device, whether the first application is a virtual machine (104);

    in response to determining that the first application is a virtual machine;

    determining (306), for the virtual machine executing on the device and by a pre-fetcher (120) using a first virtual machine physical address associated with the virtual machine, a second virtual machine physical address for data to pre-fetch for the execution of the virtual machine on the device;

    determining (308), using the second virtual machine physical address and an address mapping that associates virtual machine physical addresses for the virtual machine with device physical addresses for the device, a device physical address for the data; and

    requesting (310) the data from a memory (122) using the device physical address;

    determining (302), for a second application (110) executing on the device that is a different from the first application, whether the second application is a virtual machine; and

    in response to determining that the second application is not a virtual machine:
    determining (316), by the pre-fetcher using a second device physical address, a third device physical address for other data to pre-fetch.


     
    2. The method of claim 1, wherein requesting the data from the memory using the device physical address comprises requesting the data from a memory that is not a cache (116) of the device using the device physical address.
     
    3. The method of claim 1, comprising:
    receiving, by the pre-fetcher, the first virtual machine physical address.
     
    4. The method of claim 3, wherein receiving, by the pre-fetcher, the first virtual machine physical address comprises using, by the pre-fetcher, a cache port to retrieve the first virtual machine physical address, wherein a cache (116) includes the cache port used by the pre-fetcher and another cache port used by a processor of the device.
     
    5. The method of claim 3, wherein receiving, by the pre-fetcher, the first virtual machine physical address comprises using, by the pre-fetcher, a cache port to retrieve the first virtual machine physical address, wherein the cache port is used by the pre-fetcher and by a processor of the device to access a cache (116).
     
    6. The method of claim 3, wherein receiving, by the pre-fetcher, the virtual machine physical address comprises receiving the virtual machine physical address from a processor of the device.
     
    7. The method of claim 1, comprising:

    receiving the data from the memory in response to requesting the data from the memory using the device physical address; and

    storing (312) the data in a cache (116).


     
    8. The method of claim 1, comprising:

    receiving the data from the memory in response to requesting the data from the memory using the device physical address; and

    storing (312) the data in a buffer.


     
    9. The method of claim 8, wherein determining, for each of the multiple applications executing on the device, whether the application comprises a virtual machine comprises determining whether a register (114) on a processor of the device indicates that application comprises a virtual machine (104).
     
    10. The method of claim 9, comprising:

    detecting, for each application of the multiple applications that comprises a virtual machine (104), the execution of a launch instruction or a resume instruction to execute the virtual machine on the device; and

    in response to detecting the execution of the launch instruction or the resume instruction to execute the virtual machine on the device, setting the register to a value that indicates that the processor of the device is currently executing the virtual machine.


     
    11. The method of claim 10, comprising:

    detecting, for each application of the multiple applications that comprises a virtual machine (104), the execution of an exit instruction during the execution of the virtual machine on the device; and

    in response to detecting the execution of the exit instruction during the execution of the virtual machine on the device, setting the register to another value that indicates that the processor of the device is not currently executing the virtual machine.


     
    12. The method of claim 1, wherein:

    determining the second virtual machine physical address for the data to pre-fetch for the execution of the virtual machine on the device comprises determining the second virtual machine physical address for an instruction to pre-fetch for the execution of the virtual machine on the device; and

    requesting the data from the memory using the device physical address comprises requesting the instruction from the memory using the device physical address.


     
    13. A system (100) comprising:

    a pre-fetcher (120); and

    a non-transitory computer readable storage medium storing instructions executable by the pre-fetcher that upon such execution cause the pre-fetcher to perform operations comprising:
    determining (302), for a first application (110) executing on a device, whether the first application is a virtual machine (104);

    in response to determining that the first application is a virtual machine:

    determining (306), for the virtual machine and using a first virtual machine physical address associated with the virtual machine, a second virtual machine physical address for data to pre-fetch for the execution of the virtual machine;

    determining (308), using the second virtual machine physical address and an address mapping that associates virtual machine physical addresses for the virtual machine with device physical addresses for the device, a device physical address for the data; and requesting (310) the data from a memory (122) using the device physical address;

    determining (302), for a second application (110) that is a different application from the first application, whether the second application is a virtual machine (104); and

    in response to determining that the second application is not a virtual machine:
    determining, using a second device physical address, a third device physical address for other data to pre-fetch.
     
    14. A non-transitory computer readable storage medium storing instructions executable by a data processing apparatus and upon such execution cause the data processing to perform operations comprising:
    determining (302), for a first application (110) executing on a device and by a pre-fetcher (120), whether the first application is a virtual machine (104);
    in response to determining that the first application is a virtual machine:

    determining (306), for the virtual machine executing on the device and using a first virtual machine physical address associated with the virtual machine, a second virtual machine physical address for data to pre-fetch for the execution of the virtual machine on the device;

    determining (308), using the second virtual machine physical address and an address mapping that associates virtual machine physical addresses for the virtual machine with device physical addresses for the device, a device physical address for the data; and requesting (310) the data from a memory (122) using the device physical address;

    determining (302), for a second application (110) executing on the device that is a different application from the first application, whether the second application is a virtual machine (104); and in response to determining that the second application is not a virtual machine:
    determining (316), by the pre-fetcher using a second device physical address, a third device physical address for other data to pre-fetch.


     


    Ansprüche

    1. Verfahren, umfassend:

    Bestimmen (302) für eine erste Anwendung (110), die auf einer Vorrichtung ausgeführt wird, ob die erste Anwendung eine virtuelle Maschine (104) ist;

    als Antwort auf ein Bestimmen, dass die erste Anwendung eine virtuelle Maschine ist:

    Bestimmen (306), für die virtuelle Maschine, die auf der Vorrichtung ausgeführt wird, und durch einen Prefetcher (120) unter Verwendung einer ersten mit der virtuellen Maschine assoziierten physischen Adresse der virtuellen Maschine, einer zweiten physischen Adresse der virtuellen Maschine für Daten, die für die Ausführung der virtuellen Maschine auf der Vorrichtung zu prefetchen sind;

    Bestimmen (308), unter Verwendung der zweiten physischen Adresse der virtuellen Maschine und einer Adressabbildung, die die physischen Adressen der virtuellen Maschine für die virtuelle Maschine mit physischen Adressen der Vorrichtung für die Vorrichtung assoziiert, einer physischen Adresse der Vorrichtung für die Daten; und

    Anfordern (310) der Daten aus einem Speicher (122) unter Verwendung der physischen Adresse der Vorrichtung;

    Bestimmen (302), für eine zweite Anwendung (110), die auf der Vorrichtung ausgeführt wird, die sich von der ersten Anwendung unterscheidet, ob die zweite Anwendung eine virtuelle Maschine ist; und

    als Antwort auf ein Bestimmen, dass die zweite Anwendung keine virtuelle Maschine ist:
    Bestimmen (316), durch den Prefetcher unter Verwendung einer zweiten physischen Adresse der Vorrichtung, einer dritten physischen Adresse der Vorrichtung für andere Daten zum Prefetch.


     
    2. Verfahren gemäß Anspruch 1, wobei Anfordern der Daten aus dem Speicher unter Verwendung der physischen Adresse der Vorrichtung Anfordern der Daten aus einem Speicher umfasst, der kein Cache (116) der Vorrichtung ist, unter Verwendung der physischen Adresse der Vorrichtung.
     
    3. Verfahren gemäß Anspruch 1, umfassend:
    Empfangen, durch den Prefetcher, der ersten physischen Adresse der virtuellen Maschine.
     
    4. Verfahren gemäß Anspruch 3, wobei Empfangen, durch den Prefetcher, der ersten physischen Adresse der virtuellen Maschine, ein Verwenden eines Cache-Ports durch den Prefetcher umfasst, um die erste physische Adresse der virtuellen Maschine abzurufen, wobei ein Cache (116) den von dem Prefetcher verwendeten Cache-Port und einen anderen von einem Prozessor der Vorrichtung verwendeten Cache-Port beinhaltet.
     
    5. Verfahren gemäß Anspruch 3, wobei Empfangen, durch den Prefetcher, der ersten physischen Adresse der virtuellen Maschine, ein Verwenden eines Cache-Ports durch den Prefetcher umfasst, um die erste physische Adresse der virtuellen Maschine abzurufen, wobei der Cache-Port von dem Prefetcher und von einem Prozessor der Vorrichtung verwendet wird, um auf einen Cache (116) zuzugreifen.
     
    6. Verfahren gemäß Anspruch 3, wobei Empfangen, durch den Prefetcher, der physischen Adresse der virtuellen Maschine Empfangen der physischen Adresse der virtuellen Maschine von einem Prozessor der Vorrichtung umfasst.
     
    7. Verfahren gemäß Anspruch 1, umfassend:

    Empfangen der Daten aus dem Speicher als Antwort auf Anfordern der Daten aus dem Speicher unter Verwendung der physischen Adresse der Vorrichtung; und

    Speichern (312) der Daten in einem Cache (116).


     
    8. Verfahren gemäß Anspruch 1, umfassend:

    Empfangen der Daten aus dem Speicher als Antwort auf Anfordern der Daten aus dem Speicher unter Verwendung der physischen Adresse der Vorrichtung; und

    Speichern (312) der Daten in einem Puffer.


     
    9. Verfahren gemäß Anspruch 8, wobei Bestimmen, für jede der mehreren auf der Vorrichtung ausgeführten Anwendungen, ob die Anwendung eine virtuelle Maschine umfasst, Bestimmen umfasst, ob ein Register (114) auf einem Prozessor der Vorrichtung anzeigt, dass die Anwendung eine virtuelle Maschine (104) umfasst.
     
    10. Verfahren gemäß Anspruch 9, umfassend:

    Erfassen, für jede Anwendung der mehreren Anwendungen, die eine virtuelle Maschine (104) umfasst, der Ausführung einer Startanweisung oder einer Fortsetzungsanweisung um die virtuelle Maschine auf der Vorrichtung auszuführen; und

    als Antwort auf Erfassen der Ausführung der Startanweisung oder der Fortsetzungsanweisung um die virtuelle Maschine auf der Vorrichtung auszuführen, Einstellen des Registers auf einen Wert, der anzeigt, dass der Prozessor der Vorrichtung gerade die virtuelle Maschine ausführt.


     
    11. Verfahren gemäß Anspruch 10, umfassend:

    Erfassen, für jede Anwendung der mehreren Anwendungen, die eine virtuelle Maschine (104) umfasst, der Ausführung einer Austrittsanweisung während der Ausführung der virtuellen Maschine auf der Vorrichtung; und

    als Antwort auf Erfassen der Ausführung der Austrittsanweisung während der Ausführung der virtuellen Maschine auf der Vorrichtung, Einstellen des Registers auf einen anderen Wert, der anzeigt, dass der Prozessor der Vorrichtung die virtuelle Maschine derzeit nicht ausführt.


     
    12. Verfahren gemäß Anspruch 1, wobei:

    Bestimmen der zweiten physischen Adresse der virtuellen Maschine für die Daten, die für die Ausführung der virtuellen Maschine auf der Vorrichtung zu prefetchen sind, Bestimmen der zweiten physischen Adresse der virtuellen Maschine für eine Anweisung zum Prefetch für die Ausführung der virtuellen Maschine auf der Vorrichtung umfasst; und

    Anfordern der Daten aus dem Speicher unter Verwendung der physischen Adresse der Vorrichtung, Anfordern der Anweisung aus dem Speicher unter Verwendung der physischen Adresse der Vorrichtung umfasst.


     
    13. System (100), umfassend:

    eine Prefetcher (120); und

    ein nichtflüchtiges, computerlesbares Speichermedium, das Anweisungen speichert, die von dem Prefetcher ausführbar sind, die bei einer solchen Ausführung den Prefetcher dazu veranlassen, Operationen durchzuführen, umfassend:

    Bestimmen (302) für eine erste Anwendung (110), die auf einer Vorrichtung ausgeführt wird, ob die erste Anwendung eine virtuelle Maschine (104) ist;

    als Antwort auf Bestimmen, dass die erste Anwendung eine virtuelle Maschine ist:

    Bestimmen (306), für die virtuelle Maschine und unter Verwendung einer ersten mit der virtuellen Maschine assoziierten physischen Adresse der virtuellen Maschine, einer zweiten physischen Adresse der virtuellen Maschine für Daten zum Prefetch für die Ausführung der virtuellen Maschine;

    Bestimmen (308), unter Verwendung der zweiten physischen Adresse der virtuellen Maschine und einer Adressabbildung, die die physischen Adressen der virtuellen Maschine für die virtuelle Maschine mit den physischen Adresse der Vorrichtung für die Vorrichtung assoziiert, einer physischen Adresse der Vorrichtung für die Daten; und

    Anfordern (310) der Daten aus einem Speicher (122) unter Verwendung der physischen Adresse der Vorrichtung;

    Bestimmen (302), für eine zweite Anwendung (110) die sich von der ersten Anwendung unterscheidet, ob die zweite Anwendung eine virtuelle Maschine (104) ist; und

    als Antwort auf Bestimmen, dass die zweite Anwendung keine virtuelle Maschine ist:
    Bestimmen, unter Verwendung einer zweiten physischen Adresse der Vorrichtung, einer dritten physischen Adresse der Vorrichtung für andere Daten, die zu prefetchen sind.


     
    14. Nichtflüchtiges, computerlesbares Speichermedium, das Anweisungen speichert, die von einem Datenverarbeitungsgerät ausführbar sind, und bei einer solchen Ausführung das Datenverarbeitungsgerät veranlassen, Operationen durchzuführen, die Folgendes umfassen:

    Bestimmen (302), für eine erste Anwendung (110), die auf einer Vorrichtung ausgeführt wird, und durch einen Prefetcher (120), ob die erste Anwendung eine virtuelle Maschine (104) ist;

    als Antwort auf Bestimmen, dass die erste Anwendung eine virtuelle Maschine ist:

    Bestimmen (306), für die virtuelle Maschine, die auf der Vorrichtung ausgeführt wird, und unter Verwendung einer ersten mit der virtuellen Maschine assoziierten physischen Adresse der virtuellen Maschine, einer zweiten physischen Adresse der virtuellen Maschine für Daten, die für die Ausführung der virtuellen Maschine auf der Vorrichtung zu prefetchen sind;

    Bestimmen (308), unter Verwendung der zweiten physischen Adresse der virtuellen Maschine und einer Adressabbildung, die die physischen Adressen der virtuellen Maschine für die virtuelle Maschine mit den physischen Adresse der Vorrichtung für die Vorrichtung assoziiert, einer physischen Adresse der Vorrichtung für die Daten; und

    Anfordern (310) der Daten aus einem Speicher (122) unter Verwendung der physischen Adresse der Vorrichtung;

    Bestimmen (302), für eine zweite Anwendung (110), die auf der Vorrichtung ausgeführt wird, die sich von der ersten Anwendung unterscheidet, ob die zweite Anwendung eine virtuelle Maschine (104) ist; und als Antwort auf Bestimmen, dass die zweite Anwendung keine virtuelle Maschine ist:
    Bestimmen (316), durch den Prefetcher unter Verwendung einer zweiten physischen Adresse der Vorrichtung, einer dritten physischen Adresse der Vorrichtung für andere Daten, die zu prefetchen sind.


     


    Revendications

    1. Un procédé comprenant :

    le fait (302) de déterminer, pour une première application (110) s'exécutant sur un dispositif, si la première application est une machine virtuelle (104) ;

    en réponse à la détermination que la première application est une machine virtuelle ;

    le fait (306) de déterminer, pour la machine virtuelle s'exécutant sur le dispositif et par un pré-extracteur (120) en utilisant une première adresse physique de machine virtuelle associée à la machine virtuelle, une deuxième adresse physique de machine virtuelle pour les données à pré-extraire en vue de l'exécution de la machine virtuelle sur le dispositif ;

    le fait (308) de déterminer, en utilisant la deuxième adresse physique de machine virtuelle et un mappage d'adresses qui associe des adresses physiques de machine virtuelle pour la machine virtuelle avec des adresses physiques de dispositif pour le dispositif, une adresse physique de dispositif pour les données ; et

    le fait (310) de demander les données depuis une mémoire (122) en utilisant l'adresse physique du dispositif ;

    le fait (302) de déterminer, pour une deuxième application (110) s'exécutant sur le dispositif qui est différente de la première application, si la deuxième application est une machine virtuelle ; et

    en réponse à la détermination que la deuxième application n'est pas une machine virtuelle :
    le fait (316) de déterminer, par le pré-extracteur utilisant une deuxième adresse physique de dispositif, une troisième adresse physique de dispositif pour d'autres données à pré-extraire.


     
    2. Le procédé selon la revendication 1, dans lequel le fait de demander les données depuis la mémoire en utilisant l'adresse physique du dispositif comprend le fait de demander les données depuis une mémoire qui n'est pas un cache (116) du dispositif utilisant l'adresse physique du dispositif.
     
    3. Le procédé selon la revendication 1, comprenant :
    le fait de recevoir, par le pré-extracteur, la première adresse physique de la machine virtuelle.
     
    4. Le procédé selon la revendication 3, dans lequel le fait de recevoir, par le pré-extracteur, la première adresse physique de machine virtuelle comprend le fait d'utiliser, par le pré-extracteur, un port cache pour récupérer la première adresse physique de machine virtuelle, un cache (116) comprenant le port cache utilisé par le pré-extracteur et un autre port cache utilisé par un processeur du dispositif.
     
    5. Le procédé selon la revendication 3, dans lequel le fait de recevoir, par le pré-extracteur, la première adresse physique de machine virtuelle comprend le fait d'utiliser, par le pré-extracteur, un port cache pour récupérer la première adresse physique de machine virtuelle, le port cache étant utilisé par le pré-extracteur et par un processeur du dispositif pour accéder à un cache (116).
     
    6. Le procédé selon la revendication 3, dans lequel le fait de recevoir, par le pré-extracteur, l'adresse physique de la machine virtuelle comprend le fait de recevoir l'adresse physique de la machine virtuelle depuis un processeur du dispositif.
     
    7. Le procédé selon la revendication 1, comprenant :

    le fait de recevoir les données depuis la mémoire en réponse au fait de demander les données depuis la mémoire utilisant l'adresse physique du dispositif ; et

    le fait (312) de stocker les données dans un cache (116).


     
    8. Le procédé selon la revendication 1, comprenant :

    le fait de recevoir les données depuis la mémoire en réponse au fait de demander de données depuis la mémoire en utilisant l'adresse physique du dispositif ; et

    le fait (312) de stocker les données dans une mémoire tampon.


     
    9. Le procédé selon la revendication 8, dans lequel le fait de déterminer, pour chacune des multiples applications s'exécutant sur le dispositif, si l'application comprend une machine virtuelle comprend le fait de déterminer si un registre (114) sur un processeur du dispositif indique qu'une application comprend une machine virtuelle (104).
     
    10. Le procédé selon la revendication 9, comprenant :

    le fait de détecter, pour chaque application faisant partie des multiples applications qui comprend une machine virtuelle (104), l'exécution d'une instruction de lancement ou d'une instruction de reprise pour exécuter la machine virtuelle sur le dispositif ; et

    en réponse au fait de détecter l'exécution de l'instruction de lancement ou de l'instruction de reprise pour exécuter la machine virtuelle sur le dispositif, le fait de régler le registre à une valeur qui indique que le processeur du dispositif est en train d'exécuter la machine virtuelle.


     
    11. Le procédé selon la revendication 10, comprenant :

    le fait de détecter, pour chaque application faisant partie des multiples applications qui comprend une machine virtuelle (104), l'exécution d'une instruction de sortie pendant l'exécution de la machine virtuelle sur le dispositif ; et

    en réponse à la détection de l'exécution de l'instruction de sortie lors de l'exécution de la machine virtuelle sur le dispositif, le fait de régler le registre à une autre valeur qui indique que le processeur du dispositif n'est pas en train d'exécuter la machine virtuelle.


     
    12. Le procédé selon la revendication 1, dans lequel :

    le fait de déterminer la deuxième adresse physique de la machine virtuelle pour les données à pré-extraire pour l'exécution de la machine virtuelle sur le dispositif comprend le fait de déterminer la deuxième adresse physique de la machine virtuelle pour une instruction de pré-extraire pour l'exécution de la machine virtuelle sur le dispositif ; et

    le fait de demander les données depuis la mémoire en utilisant l'adresse physique du dispositif comprend le fait de demander l'instruction depuis la mémoire en utilisant l'adresse physique du dispositif.


     
    13. Un système (100) comprenant :

    un pré-extracteur (120) ; et

    un support de stockage lisible par ordinateur non transitoire stockant des instructions exécutables par le pré-extracteur qui, lors d'une telle exécution, amène le pré-extracteur à effectuer des opérations comprenant :

    le fait (302) de déterminer, pour une première application (110) s'exécutant sur un dispositif, si la première application est une machine virtuelle (104) ;

    en réponse à la détermination que la première application est une machine virtuelle :

    le fait (306) de déterminer, pour la machine virtuelle et en utilisant une première adresse physique de machine virtuelle associée à la machine virtuelle, une deuxième adresse physique de machine virtuelle pour les données à pré-extraire en vue de l'exécution de la machine virtuelle ;

    le fait (308) de déterminer, en utilisant la deuxième adresse physique de machine virtuelle et un mappage d'adresses qui associe des adresses physiques de machine virtuelle pour la machine virtuelle à des adresses physiques de dispositif pour le dispositif, une adresse physique de dispositif pour les données ; et

    le fait de demander (310) les données depuis une mémoire (122) en utilisant l'adresse physique du dispositif ;

    le fait (302) de déterminer, pour une deuxième application (110) qui est une application différente de la première application, si la deuxième application est une machine virtuelle (104) ; et

    en réponse à la détermination que la deuxième application n'est pas une machine virtuelle :
    le fait de déterminer, en utilisant une deuxième adresse physique de dispositif, une troisième adresse physique de dispositif pour d'autres données à pré-extraire.


     
    14. Un support de stockage lisible par ordinateur non transitoire stockant des instructions exécutables par un appareil de traitement de données et, lors d'une telle exécution, amenant le traitement de données à effectuer des opérations comprenant :

    le fait (302) de déterminer, pour une première application (110) s'exécutant sur un dispositif et par un pré-extracteur (120), si la première application est une machine virtuelle (104) ;

    en réponse à la détermination que la première application est une machine virtuelle :

    le fait (306) de déterminer, pour la machine virtuelle s'exécutant sur le dispositif et utilisant une première adresse physique de machine virtuelle associée à la machine virtuelle, une deuxième adresse physique de machine virtuelle pour des données à pré-extraire en vue de l'exécution de la machine virtuelle sur le dispositif ;

    le fait (308) de déterminer, en utilisant la deuxième adresse physique de machine virtuelle et un mappage d'adresses qui associe des adresses physiques de machine virtuelle pour la machine virtuelle avec des adresses physiques de dispositif pour le dispositif, une adresse physique de dispositif pour les données ; et le fait (310) de demander les données depuis une mémoire (122) en utilisant l'adresse physique du dispositif ;

    le fait (302) de déterminer, pour une deuxième application (110) s'exécutant sur le dispositif qui est une application différente de la première application, si la deuxième application est une machine virtuelle (104) ; et en réponse à la détermination que la deuxième application n'est pas une machine virtuelle :
    le fait (316) de déterminer, par le pré-extracteur en utilisant une deuxième adresse physique de dispositif, une troisième adresse physique de dispositif pour d'autres données à pré-extraire.


     




    Drawing















    REFERENCES CITED IN THE DESCRIPTION



    This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

    Patent documents cited in the description