WO2006081582A2 - A method and apparatus for supporting address translation in a virtual machine environment - Google Patents

A method and apparatus for supporting address translation in a virtual machine environment Download PDF

Info

Publication number
WO2006081582A2
WO2006081582A2 PCT/US2006/003587 US2006003587W WO2006081582A2 WO 2006081582 A2 WO2006081582 A2 WO 2006081582A2 US 2006003587 W US2006003587 W US 2006003587W WO 2006081582 A2 WO2006081582 A2 WO 2006081582A2
Authority
WO
WIPO (PCT)
Prior art keywords
data structure
translation data
shadow
active
guest
Prior art date
Application number
PCT/US2006/003587
Other languages
French (fr)
Other versions
WO2006081582A3 (en
Inventor
Andrew Anderson
Original Assignee
Intel Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Intel Corporation filed Critical Intel Corporation
Priority to CN2006800031107A priority Critical patent/CN101107593B/en
Priority to EP06720101.2A priority patent/EP1856606B1/en
Priority to JP2007553378A priority patent/JP5038907B2/en
Priority to KR1020077017491A priority patent/KR100928353B1/en
Publication of WO2006081582A2 publication Critical patent/WO2006081582A2/en
Publication of WO2006081582A3 publication Critical patent/WO2006081582A3/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02Addressing or allocation; Relocation
    • G06F12/08Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/10Address translation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02Addressing or allocation; Relocation
    • G06F12/08Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/10Address translation
    • G06F12/1027Address translation using associative or pseudo-associative address translation means, e.g. translation look-aside buffer [TLB]
    • G06F12/1036Address translation using associative or pseudo-associative address translation means, e.g. translation look-aside buffer [TLB] for multiple virtual address spaces, e.g. segmentation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • G06F2009/45583Memory management, e.g. access or allocation

Definitions

  • Embodiments of the invention relate generally to virtual machines, and more specifically to supporting address translation in a virtual machine environment.
  • a conventional virtual-machine monitor typically runs on a computer and presents to other software the abstraction of one or more virtual machines.
  • Each virtual machine may function as a self-contained platform, running its own “guest operating system” (i.e., an operating system (OS) hosted by the VMM) and other software, collectively referred to as guest software.
  • guest operating system i.e., an operating system (OS) hosted by the VMM
  • OS operating system
  • guest software collectively referred to as guest software.
  • the guest software expects to operate as if it were running on a dedicated computer rather than a virtual machine. That is, the guest software expects to control various events and have access to hardware resources such as physical memory and memory-mapped input/output (I/O) devices.
  • I/O input/output
  • the guest software expects to maintain control over address-translation operations and have the ability to allocate physical memory, provide protection from and between guest applications, use a variety of paging techniques, etc.
  • the VMM should be able to have ultimate control over the computer's resources to provide protection from and between virtual machines.
  • Figure 1 illustrates one embodiment of a virtual-machine environment, in which the present invention may operate
  • Figure 2 illustrates operation of a virtual TLB, according to one embodiment of the present invention
  • Figures 3A and 3B illustrate a process of creating and maintaining metadata for a shadow PT hierarchy, according to two alternative embodiments of the present invention
  • Figure 4 is a flow diagram of one embodiment of a process for synchronizing guest translation data structure and shadow translation data structure
  • Figure 5 is a flow diagram of one embodiment of a process for maintaining metadata for a shadow translation data structure
  • Figure 6 is a flow diagram of one embodiment of a process for facilitating a change of an address space
  • Figure 7 is a flow diagram of one embodiment of a process for synchronizing entries of two translation data structures for a specified address
  • Figure 8 is a flow diagram of one embodiment of a process for removing a shadow PT hierarchy from a working set of shadow PT hierarchies maintained by the VMM;
  • Figure 9 is a flow diagram of one embodiment of a process for adding an entry to a PD of a shadow PT hierarchy
  • Figure 10 is a flow diagram of one embodiment of a process for removing an entry from a PD of a shadow PT hierarchy
  • Figure 11 is a flow diagram of one embodiment of a process for adding an entry to a PT of a shadow PT hierarchy
  • Figure 12 is a flow diagram of one embodiment of a process for removing an entry from a PT of a shadow PT hierarchy
  • Figure 13 is a flow diagram of one embodiment of a process for monitoring a PTE of a shadow PT hierarchy
  • Figure 14 is a flow diagram of one embodiment of a process for removing monitoring from a PTE of a shadow PT hierarchy.
  • the present invention may be provided as a computer program product or software which may include a machine or computer- readable medium having stored thereon instructions which may be used to program a computer (or other electronic devices) to perform a process according to the present invention.
  • processes of the present invention might be performed by specific hardware components that contain hardwired logic for performing the processes, or by any combination of programmed computer components and custom hardware components.
  • a machine-readable medium may include any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer), but is not limited to, floppy diskettes, optical disks, Compact Disc, Read-Only Memories (CD- ROMs), and magneto-optical disks, Read-Only Memories (ROMs), Random Access Memories (RAMs), Erasable Programmable Read-Only Memories (EPROMs), Electrically Erasable Programmable Read-Only Memories (EEPROMs), magnetic or optical cards, flash memories, a transmission over the Internet, electrical, optical, acoustical or other forms of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc.) or the like.
  • a machine e.g., a computer
  • ROMs Read-Only Memories
  • RAMs Random Access Memories
  • EPROMs Erasable Programmable Read-Only Memories
  • EEPROMs Electrically Erasable Programmable
  • a design may go through various stages, from creation to simulation to fabrication.
  • Data representing a design may represent the design in a number of manners.
  • the hardware may be represented using a hardware description language or another functional description language.
  • a circuit level model with logic and/or transistor gates may be produced at some stages of the design process.
  • most designs, at some stage reach a level of data representing the physical placement of various devices in the hardware model.
  • data representing a hardware model may be the data specifying the presence or absence of various features on different mask layers for masks used to produce the integrated circuit.
  • the data may be stored in any form of a machine- readable medium.
  • An optical or electrical wave modulated or otherwise generated to transmit such information, a memory, or a magnetic or optical storage such as a disc may be the machine readable medium. Any of these mediums may "carry” or “indicate” the design or software information.
  • an electrical carrier wave indicating or carrying the code or design is transmitted, to the extent that copying, buffering, or re-transmission of the electrical signal is performed, a new copy is made.
  • a communication provider or a network provider may make copies of an article (a carrier wave) embodying techniques of the present invention.
  • FIG. 1 illustrates one embodiment of a virtual-machine environment 100, in which the present invention may operate.
  • bare platform hardware 116 comprises a computing platform, which may be capable, for example, of executing a standard operating system (OS) or a virtual-machine monitor (VMM), such as a VMM 112.
  • OS operating system
  • VMM virtual-machine monitor
  • the VMM 112 may emulate and export a bare machine interface to higher level software.
  • higher level software may comprise a standard or real-time OS, may be a highly stripped-down operating environment with limited operating system functionality, may not include traditional OS facilities, etc.
  • the VMM 112 may be run within, or on top of, another VMM.
  • VMMs may be implemented, for example, in hardware, software, firmware or by a combination of various techniques.
  • the platform hardware 116 can be of a personal computer (PC), mainframe, handheld device, portable computer, set-top box, or any other computing system.
  • the platform hardware 116 includes a processor 118 and memory 120.
  • Processor 118 can be any type of processor capable of executing software, such as a microprocessor, digital signal processor, microcontroller, or the like.
  • the processor 118 may include microcode, programmable logic or hardcoded logic for performing the execution of method embodiments of the present invention. Although Figure 1 shows only one such processor 118, there may be one or more processors in the system.
  • Memory 120 can be a hard disk, a floppy disk, random access memory (RAM) (e.g., dynamic RAM (DRAM) or static RAM (SRAM)), read only memory (ROM), flash memory, any combination of the above devices, or any other type of machine medium readable by processor 118.
  • RAM random access memory
  • ROM read only memory
  • Memory 120 may store instructions and/or data for performing the execution of method embodiments of the present invention.
  • the VMM 112 presents to other software (i.e., "guest" software) the abstraction of one or more virtual machines (VMs), which may provide the same or different abstractions to the various guests.
  • Figure 1 shows two VMs, 102 and 114.
  • the guest software running on each VM may include a guest OS such as a guest OS 104 or 106 and various guest software applications 108 and 110.
  • a guest OS such as a guest OS 104 or 106 and various guest software applications 108 and 110.
  • Each of the guest OSs 104 and 106 expects to access physical resources (e.g., processor registers, memory and I/O devices) within the VMs 102 and 114 on which the guest OS 104 or 106 is running and to perform other functions.
  • the guest OS 104 or 106 expects to have access to all registers, caches, structures, I/O devices, memory and the like, according to the architecture of the processor and platform presented in the VM 102 and 114.
  • the resources that can be accessed by the guest software may either be classified as "privileged” or "non-privileged.”
  • privileged resources the VMM 112 facilitates functionality desired by guest software while retaining ultimate control over these privileged resources.
  • Non-privileged resources do not need to be controlled by the VMM 112 and can be accessed directly by guest software.
  • each guest OS expects to handle various fault events such as exceptions (e.g., page faults, general protection faults, etc.), interrupts (e.g., hardware interrupts, software interrupts), and platform events (e.g., initialization (INIT) and system management interrupts (SMIs)).
  • exceptions e.g., page faults, general protection faults, etc.
  • interrupts e.g., hardware interrupts, software interrupts
  • platform events e.g., initialization (INIT) and system management interrupts (SMIs)
  • IIT initialization
  • SMIs system management interrupts
  • control may be transferred to the VMM 112.
  • the transfer of control from guest software to the VMM 112 is referred to herein as a VM exit.
  • the VMM 112 may return control to guest software.
  • the transfer of control from the VMM 112 to guest software is referred to as a VM entry.
  • the processor 118 controls the operation of the VMs 102 and 114 in accordance with data stored in a virtual machine control structure (VMCS) 125.
  • the VMCS 125 is a structure that may contain state of guest software, state of the VMM 112, execution control information indicating how the VMM 112 wishes to control operation of guest software, information controlling transitions between the VMM 112 and a VM, etc.
  • the processor 118 reads information from the VMCS 125 to determine the execution environment of the VM and to constrain its behavior.
  • the ViVICS is stored in memory 120.
  • multiple VMCS structures are used to support multiple VMs.
  • an address translation mechanism expected by an OS may be based on a translation lookaside buffer (TLB) 122 controlled by the processor 118 and a translation data structure, such as a page-table (PT) hierarchy, controlled by the OS and used to translate virtual memory addresses into physical memory addresses when paging is enabled.
  • TLB translation lookaside buffer
  • PT page-table
  • the architecture of the Intel ® Pentium ® 4 Processor supports a number of paging modes.
  • the most commonly used paging mode supports a 32-bit linear address space using a two-level hierarchical paging structure (referred to herein as a two-level hierarchy paging mode).
  • Embodiments of the invention are not limited to this paging mode, but instead may be employed by one skilled in the art to virtualize other paging modes (e.g., Physical Address Extension (PAE) mode, Intel® Extended Memory 64 Technology (EM64T) mode, etc.) and implementations (e.g., hashed page tables).
  • PEE Physical Address Extension
  • E64T Intel® Extended Memory 64 Technology
  • translation of a virtual memory address into a physical memory address begins with searching the TLB 122 using either the upper 20 bits (for a 4 KB page frame) or the upper 10 bits (for a 4MB page frame) of the virtual address. If a match is found (a TLB hit), the upper bits of a physical page frame that are contained in the TLB 122 are conjoined with the lower bits of the virtual address to form a physical address.
  • the TLB also contains access and permission attributes associated with the mapping. If no match is found (a TLB miss), the processor consults the PT hierarchy to determine the virtual-to-physical translation, which is then cached in the TLB 122. Entries in the PT hierarchy may include some attributes that are automatically set by the processor on certain accesses.
  • the TLB 122 may become inconsistent with the PT hierarchy if a corresponding address translation exists in the TLB 122.
  • the OS may expect to be able to resolve such an inconsistency by issuing an instruction to the processor 118.
  • ISA instruction set architecture
  • a processor allows software to invalidate cached translations in the TLB by issuing the INVLPG instruction.
  • the OS may expect to request the processor 118 to change the address space completely, which should result in the removal of all translations from the TLB 122.
  • an OS may use a MOV instruction or a task switch to request a processor to load CR3 (which contains the base address of the PT hierarchy), thereby removing all translations from the TLB.
  • CR3 which contains the base address of the PT hierarchy
  • Different levels of the page table hierarchy may have different names based upon mode and implementation.
  • the CR3 register points to the base of the page directory page. Entries in the page directory may either specify a mapping to a large-size page (e.g., a 4MB superpage, a 2MB superpage, IGB superpage, etc.), or a reference to a page table.
  • the page table in turn may contain mappings to small-size pages.
  • the VMM 112 should be able to have ultimate control over physical resources including the TLB 122.
  • Embodiments of the present invention address the conflict between the expectations of the VMs 102 and 114 and the role of the VMM 112 by using a virtual TLB that emulates the functionality of the processor's physical TLB.
  • the virtual TLB includes the TLB 122 and a set of shadow PT hierarchies controlled by the VMM 112.
  • the set of shadow PT hierarchies derive its format and content from guest PT hierarchies that may be currently used or not used by the VM 102 or 114. If the VM 102 or 114 modifies the content of the guest PT hierarchies, this content becomes inconsistent with the content of the shadow PT hierarchies.
  • the inconsistencies between the guest PT hierarchies and the shadow PT hierarchies are resolved using techniques analogous to those employed by the processor 118 in managing the TLB 122.
  • the VMM evaluates the event and synchronizes all maintained shadow PT hierarchies with the current guest state if needed.
  • the same shadow PT may be a part of multiple guest PT hierarchies.
  • the corresponding shadow PT will in turn be a member of multiple shadow PT hierarchies.
  • synchronization performed by the VMM may update shadow page table or page directory entries for a shadow PT hierarchy that is not currently in-use. Likewise synchronization may be required to guest pages that are not part of the in-use guest PT hierarchy.
  • the VMM 112 includes an address translation module 126 that is responsible for creating and maintaining a working set of shadow PT hierarchies for each of the VM 102 and 114 in a virtual TLB (VTLB) data store 124.
  • the working set of shadow PT hierarchies is maintained for corresponding active processes of the VM 102 or 114 (i.e., processes that are likely to be activated in the near future by the VM 102 or 114).
  • active processes of the VM 102 or 114 i.e., processes that are likely to be activated in the near future by the VM 102 or 114.
  • the IA32 ISA the only explicitly defined guest hierarchy is that defined by the currently used paging structures. In practice there is a high deal of temporal locality for guest processes and their address spaces.
  • the VMM may employ heuristics or explicit information to determine a set of active process.
  • the address translation module 126 identifies a corresponding shadow PT hierarchy in the working set and requests the processor 118 to load its base address. When applicable, the address translation module 126 can then reuse previously computed mappings that are stored in the shadow PT hierarchies.
  • the address translation module 126 derives a new shadow PT hierarchy from a corresponding guest PT hierarchy and adds it to the working set. Alternatively, if the VM 102 or 114 de-activates an existing process, the address translation module 126 removes information corresponding to the guest PT hierarchy from the working set.
  • the address translation module 126 is responsible for extracting metadata from each new shadow PT hierarchy, storing the metadata in the VTLB data store 124, and updating the metadata when the shadow PT hierarchy is modified.
  • the metadata includes a PT vector (PTV), a PD vector (PDV), an active PTE list, and an active PDE list.
  • the PTV and PDV track the guest frames that are used as PTs and PDs. In one embodiment, this information is encoded in bit vectors.
  • the PTV may be indexed by page frame number (PFN), with each entry bit being set if a corresponding PFN is a PT.
  • the PDV may be indexed by a page frame number (PFN), with each entry bit being set if a corresponding PFN is a PD.
  • the active PTE list is a list of PT entries (PTEs) in the shadow PT hierarchy that point to frames holding PTs and PD.
  • the active PDE list identifies PD entries (PDEs) in the shadow PT hierarchy that point to PTs containing PT entries identified in the active PTE list.
  • active PDE and PTE lists contain additional metadata describing whether the mapping is to a PD or PT frame.
  • embodiments of this invention may use a variety of data structures which may be more or less space or time efficient than those described herein.
  • One skilled in the art will also recognize the extension of tracking structures to support additional paging modes. For example, an EM64T paging mode maps a 64-bit virtual address to a physical address through a four-level hierarchical paging structure. The actual number of bits supported in the virtual or physical address spaces may be implementation dependent and may be less than 64 bits in a particular implementation.
  • an EM64T implementation may require additions of a page-map level 4 (PML4) page vector and a page directory pointer (PDP) page vector to track the additional page tables used in the EM64T paging structure.
  • PML4 page-map level 4
  • PDP page directory pointer
  • the active PTE list will be extended to include entries which map any page used within the paging structures (e.g., PML4 or PDP pages for EM64T).
  • active PTE/PDE list metadata is maintained to track the number of PD and PT frames that are mapped through a page table.
  • the number of mappings per page is incremented from 0, then PDEs which map the PT must be added to the active PDE list, and when the number of mappings is decreased to zero, then PDEs that map this PT must be removed from the active PDE list.
  • the address translation module 126 is responsible for synchronizing a current shadow PT hierarchy with a current guest PT hierarchy when such synchronization is needed.
  • the address translation module 126 performs the synchronization by determining which entries in the guest PT hierarchy have recently been modified and then updating corresponding entries in the shadow PT hierarchy accordingly.
  • the address translation module 126 determines which entries in the guest PT hierarchy have recently been modified based on the metadata extracted from the shadow PT hierarchy and attributes associated with the entries of the shadow PT hierarchy.
  • the attributes include access attributes associated with PD entries in the shadow PT hierarchy and update attributes associated with PT entries in the shadow PT hierarchy.
  • Figure 2 illustrates operation of a virtual TLB 204, according to one embodiment of the present invention.
  • Virtual TLB 204 includes a shadow translation data structure represented by a shadow PT hierarchy 206 and a physical TLB 208.
  • the shadow PT hierarchy 206 derives its structure and content from a guest translation data structure represented by a guest PT hierarchy 202.
  • the VMM maintains a working set of shadow PT hierarchies for active processes of the VM.
  • the VM when the VM requests the processor to enable a different guest PT hierarchy (e.g., by issuing MOV to CR3 or task switch in the IA-32 ISA), control transitions to the VMM, which instructs the processor to load the base address 214 of a shadow PT hierarchy 206 corresponding to the requested guest PT hierarchy 202.
  • this shadow PT hierarchy 206 is synchronized with the guest PT hierarchy 202 using relevant metadata and attributes, as will be discussed in greater detail below.
  • the virtual TLB maintains access and update attributes in the entries of the shadow PD and PTs. These attributes are also referred to as an accessed (A) bit and a dirty (D) bit.
  • the processor sets the accessed (A) attribute in the corresponding PT entry or PD entry in the shadow PT hierarchy 206. If guest software attempts to write a page frame, the processor sets the dirty (D) attribute in the corresponding shadow PT entry.
  • Guest software is allowed to freely modify the guest PT hierarchy 202 including changing virtual-to-physical mapping, permissions, etc. Accordingly, the shadow PT hierarchy 206 may not be always consistent with the guest PT hierarchy 202.
  • the guest OS which treats the virtual TLB 204 as a physical TLB, attempts to change the virtual TLB 204 by requesting a processor to perform an operation defined by a relevant ISA.
  • such operations include the INVLPG instruction, CR3 loads, paging activation (modification of CR0.PG), modification of global paging (toggling of the CR4.PGE bit), etc.
  • the operations attempting to change the virtual TLB 204 are configured by the VMM as privileged (e.g., using corresponding execution controls stored in the VMCS), and, therefore, result in a VM exit to the VMM.
  • the VMM determines the cause of the VM exit and modifies the content of the shadow PT hierarchy 206 if necessary. For example, if the VM exit occurs due to a page fault that should be handled by the guest OS (e.g., a page fault caused by an access not permitted by the guest PT hierarchy 202), the page fault is injected to the guest OS for handling.
  • FIG. 3 A illustrates a process of creating and maintaining metadata for a shadow PT hierarchy in a two-level hierarchy paging mode, according to one embodiment of the present invention.
  • FIG. 3 A a number of physical page frames identified by distinct letters (letters A through W) is illustrated.
  • Some guest page frames may contain a PD (e.g., frame A).
  • Other guest page frames may contain a PT (e.g., frames A, B, C, and L).
  • a hierarchy 302 is a guest PT hierarchy.
  • FIG. 3 A shows a shadow PT hierarchy 304 created based on a guest PT hierarchy 302.
  • Each PD or PT in the guest PT hierarchy 302 includes a corresponding PD or PT in the shadow PT hierarchy 304.
  • a shadow page is not required for each page in the guest PT.
  • Some embodiments may choose to restrict shadow pages according to usage statistics (e.g., only generate shadow pages for guest PT pages that have been used), or according to resource constraints (e.g., maintaining only a set of shadow pages based on available memory).
  • Separate shadow tables are maintained for PD and PT tables derived from the same physical frame. For example, separate tables 330 and 332 are maintained for PD 306 and PT 308 that are derived from the same physical frame 314.
  • the PD and PT entries in the shadow PT hierarchy 304 contain transformed mappings for the guest frames 314 through 324.
  • frames 316 and 318 are used as PTs 310 and 312, and frame 314 is used both as PD 306 and PT 308. This usage is illustrated as "PT” and "PD/PT" in the page frames 314 through 316 shown under the shadow PT hierarchy 304.
  • the shadow PT hierarchy 304 is associated with an active PTE list 342 and an active PDE list 344.
  • the active PTE list 342 identifies PT entries in the shadow PT hierarchy 304 that map PT and PD page frames from the guest PT hierarchy 302.
  • the active PTE list 342 identifies entries in the PT 332 that map page frames 314 through 318.
  • the active PDE list 344 identifies PD entries in the shadow PT hierarchy that point to PTs with entries identified in the active PTE list 342.
  • the active PDE list 344 includes entries in the PD 330 that point to the PT 332.
  • the active PTE list 342 and the active PDE list 344 are components of the metadata of the shadow PT hierarchy 304.
  • the shadow PT hierarchy 304 is associated with a PT bit vector (PTV) 362 and a PD bit vector (PDV) 364.
  • the PTV 362 tracks the guest page frames that are used as PTs.
  • the PTV 362 includes page frames 314 through 318 which are used as PTs in the guest PT hierarchy 302.
  • the PDV 364 tracks the guest page frames that are used as PDs.
  • the PDV 364 includes page frame 314 that is used as PD in the guest PT hierarchy 302.
  • the PTV 362 and PDV 364 represent all shadow PT hierarchies in the working set and track the capacity in which shadow pages are employed in the working set (e.g., if a shadow page has not been allocated for a guest PT, then the PTV will not reflect the guest PT page, even if it appears in the guest paging structures).
  • the VMM may detect this addition (e.g., on the next or subsequent VM exit related to TLB manipulation) and add a corresponding PT to the shadow PT hierarchy 304. For example, if a new PT 352 derived from a frame 319 is added to the guest PT hierarchy 302, with a mapping for a new frame 354, the VMM may add a corresponding PT 360 with transformed mappings to the shadow PT hierarchy 304 and update the metadata to reflect this change.
  • the VMM adds an entry mapping frame 319 in the PT 332 to the active PTE list 342, and an entry pointing to the PT 360 in the PD 330 to the active PDE list 344. Also, the VMM adds frame 319 to PTV 362, which tracks guest frames (i.e., here frame 319) used as PTs.
  • Figure 3B illustrates a process of creating and maintaining metadata for a shadow PT hierarchy in the EM64T paging mode, according to one embodiment of the present invention.
  • the base of the paging structure is a PML4 page (e.g., frame A).
  • PML4 page e.g., frame A
  • PDP page e.g., frames B and C
  • PDP page e.g., frames B and C
  • PDP page e.g., frames D and E
  • PD page directory
  • PT page table
  • Each PML4, PDP, PD or PT page may be 4KB in size.
  • the entry size may be increased relative to the 32-bit paging mode. Specifically, there may be 512 entries per page, requiring that 9 bits of the virtual address be used at each level to select the appropriate entry. This selector size may lead to a large page size of 2MB instead of 4MB as described previously.
  • the creation of metadata in the EM64T paging mode includes the generation of several vectors, an active entry list, and several active directory lists.
  • the vectors include a PML4V vector identifying frames used as PML4 pages, a
  • the active entry list is an active PTE list including all mappings which map a PML4, PDP, PD or PT page.
  • the active directory lists include lists identifying higher level mapping structures referencing a lower level structure through which the guest page corresponding to a shadow structure can be accessed.
  • the active directory lists consist of an active PDE list including those PDEs that reference a page containing active PTE list entries, an active PDPE list including active PDPE entries which reference a PD containing an active PDE list entry, and an active PML4E list including entries which map a PDP containing elements in the active PDPE list.
  • the synchronization of the shadow page tables begins with checking each entry in the active PML4E list associated with the used shadow PT hierarchy. If the entry has been accessed, each element in the active PDPE list corresponding to the accessed PML4 entry is checked, and then the processing continues as previously described.
  • active lists are not maintained and/or processed for one or more of the upper levels of the hierarchy.
  • the use of an active list for each level of the hierarchy will cause this single entry to be always accessed, thereby allowing no reduction in the amount of processing required for lower levels in the hierarchy.
  • the synchronization may instead begin by processing an active list lower in the hierarchy.
  • active PDPE list elements may first be processed followed by active PDE list elements or active ⁇ PTE list elements associated with a used shadow PT hierarchy.
  • the initial layer processed on synchronization may be predetermined.
  • the initial layer to be processed may be determined by dynamic profiling of the guest's page table usage.
  • IA-32 supports an additional paging mode in which a 32-bit virtual address is mapped to a larger physical address.
  • the page table base register is configured to point to a PDP page which contains four elements. Entry sizes and behaviors in this additional mode of operations are similar to those described above for the 64-bit virtual address mode. As this additional mode does not make use of PML4 pages, the PML4V and active PML4E list are not required.
  • FIG. 4 is a flow diagram of one embodiment of a process 400 for synchronizing a guest translation data structure and a shadow translation data structure.
  • the process may be performed by processing logic that may comprise hardware (e.g., dedicated logic, programmable logic, microcode, etc.), software (such as that run on a general purpose computer system or a dedicated machine), or a combination of both.
  • the process is performed by an address translation module 126 of Figure 1.
  • process 400 begins with processing logic receiving control transitioned from a VM due to an event pertaining to manipulation of the TLB (processing block 402).
  • events may include a request to change the current address space (e.g., CR3 load), a request to adjust inconsistent translations for a specified virtual address in the TLB (e.g., INVLPG), a page fault, etc.
  • processing logic determines whether the event pertaining to the manipulation of the TLB should be handled by the VM. If so (e.g., the event is a page fault caused by a problematic mapping in a guest translation data structure), control is returned to the VM for handling the event (processing block 406). If not, processing logic determines whether the event is associated with a specified problematic address (processing box 408). [0071] If the event does not need to be handled by the VM, it may be associated with a specified problematic address.
  • Examples of such an event may include an event caused by the INVLPG instruction that takes a problematic address as an operand, an event caused by an induced page fault (e.g., a page fault resulting from an inconsistency between the two translation data structures with respect to a specific mapping, a page fault caused by a need to virtualize A/D bits in the guest translation data structure, etc.), etc.
  • processing logic makes corrections in the shadow translation data structure for the specified address (e.g., removes a stale mapping for the specified address or adds a new mapping for the specified address) to conform to the guest translation data structure (processing block 410).
  • One embodiment of a process for synchronizing entries of two translation data structures for a specified address is discussed in more detail below in conjunction with Figure 7.
  • processing logic determines which entries of the guest translation data structure have been modified (processing block 412). The determination is made using metadata extracted from the shadow translation data structure and attributes associated with the entries of the shadow translation data structure (processing block 412).
  • the metadata includes vectors and active lists for various levels of the shadow translation data structure. A vector for a specific level of the shadow translation data structure identifies frames used as pages at this level of the guest translation data structure.
  • the active lists include an active entry list and one or more active directory lists.
  • the active entry list includes mappings that map pages used by the guest in forming the guest translation data structure.
  • the active directory lists identify higher level mapping structures referencing a lower level structure through which a guest page corresponding to a shadowed paging structure can be accessed.
  • the metadata includes, in one embodiment, vectors PTV and PDV, an active entry list (a PTE list), and an active directory list (a PDE list).
  • the metadata includes, in one embodiment, vectors PTV, PDV, PDPV and PML4V, an active entry list (a PTE list), and active directory lists (an active PDE list, an active PDPE list and an active PML4E list).
  • processing logic synchronizes corresponding entries in the shadow translation data structure with the modified entries of the guest translation data structure. Accordingly, processing logic only needs to synchronize the entries that were modified, rather than re-populating the entire content of the shadow translation data structure.
  • extra storage is used to maintain some guest PD and/or PT contents as they were last synchronized. This permits the VMM to determine where modifications have been made without calculating or looking up additional relocation or permission information.
  • Figures 5-14 illustrate various processes performed to support address translation in a virtual machine environment using the two-level hierarchy paging mode, according to different embodiments of the present invention. These processes may be performed by processing logic that may comprise hardware (e.g., dedicated logic, programmable logic, microcode, etc.), software (such as that run on a general purpose computer system or a dedicated machine), or a combination of both. In one embodiment, each of these processes is performed by an address translation module 126 of Figure 1.
  • processing logic may comprise hardware (e.g., dedicated logic, programmable logic, microcode, etc.), software (such as that run on a general purpose computer system or a dedicated machine), or a combination of both.
  • each of these processes is performed by an address translation module 126 of Figure 1.
  • Figure 5 is a flow diagram of one embodiment of a process 500 for maintaining metadata for a shadow translation data structure such as a shadow PT hierarchy.
  • process 500 begins with processing logic creating a shadow page for each PD or PT page from the guest PT hierarchy (processing block 502).
  • processing logic tracks page frames used as PDs or PTs in the guest PT hierarchy. In one embodiment, processing logic sets an entry in the PDV if a corresponding PFN is a PD in the guest PT hierarchy. Similarly, processing logic sets an entry in the PTV if a corresponding PFN is a PT in the guest PT hierarchy.
  • processing logic tracks mappings to any Dynamic Random Access Memory (DRAM) backed page (to identify pages that can potentially be PDs or PTs).
  • processing logic tracks mappings to DRAM based pages using an inverted page table (IPT) and an inverted page directory (IPD).
  • IPT is indexed by a PFN of a data page frame, with each entry containing a list of addresses of PTEs that map the data page frame.
  • IPD is indexed by a PFN of the page table, with each entry containing a list of addresses of PDEs that reference the PFN as a page table.
  • processing logic identifies 4MB pages in the guest PT hierarchy and creates a page table in the shadow PT hierarchy for each 4MB page to avoid large page mappings and thereby reduce future synchronization time. Otherwise, an update of a 4MB page would cause the synchronization of every PD and PT page within the 4MB.
  • an inverted expansion table IET is used to track which PDEs in the guest PT hierarchy point to a 4MB page. The IET is indexed by a PFN and attribute bits, with every entry listing PDEs that point to the exploded 4MB page.
  • the IPD may be indexed by the address of the shadow PFN to minimize required address translation steps.
  • memory type information (e.g., cacheability information) can be stored in PAT bits within the PDE/PTE that maps a page. This type information is not captured in a PDE that is a page-table pointer. Hence, if two 4MB pages were to map the same region with different PAT attributes, then separate page tables would be required to convey the correct PAT attributes. Using separate expansion tables for each set of attributes resolves this issue.
  • processing logic identifies PTEs in the shadow PT hierarchy that map pages used as PD or PT in the guest PT hierarchy and creates an active PTE list. [0086] At processing block 512, processing logic identifies PDEs in the shadow
  • processing logic changes the above active PTE and PDE lists accordingly.
  • FIG. 6 is a flow diagram of one embodiment of a process 600 for facilitating a change of an address space. Note that in IA32 the same CR3 value may also be reloaded to force a flush of stale TLB mappings. Similar processing steps are taken for a change of CR3 or for a CR3 reload. [0089] Referring to Figure 6, process 600 begins with processing logic determining that a VM exit occurred due to a request of the VM to enable a different guest PT hierarchy (e.g., by issuing a CR3 load request) (processing block 602).
  • processing logic scans all active PDEs corresponding to the currently in-use shadow PT hierarchy identified in the active PDE list of the metadata to find which of these PDEs have been accessed (have an access attribute set to an access value) (processing block 604), and then initializes the access attributes of the accessed PDEs (processing block 606).
  • n ⁇ n-leaf paging tables do not support a dirty bit. If the accessed bit is clear, then no page within the 4MB region has been read or written, so any guest page table or page directory cannot have been modified. However, the accessed bit does not distinguish between reads and writes, so 4MB regions which have been accessed should be further processed even though it is possible that nothing has been modified. In architectures supporting a dirty bit for non-leaf page tables, the dirty bit is checked instead, and only regions which had been written to require further processing.
  • processing logic scans all shadow PTEs corresponding to the accessed active PDE in the active PTE list of the metadata to find which of these PTEs include mappings for an updated page (have an update attribute set to an update value) (processing block 608).
  • processing logic compares PD/PT entries in the guest PT hierarchy with corresponding entries in the shadow PT hierarchy (processing block 610) and changes the corresponding entries of the shadow PT hierarchy to conform to the modified entries of the guest PT hierarchy (e.g., by removing from the shadow PT hierarchy a PTE/PDE absent in the guest PT hierarchy, by adding to the shadow PT hierarchy a new PTE/PDE recently added to the guest PT hierarchy, etc.) (processing block 612). Note that adding PDEs may require the allocation and initialization of additional shadow PTs. This in turn may require updates to the various metadata structures maintained by the address translation module 126. [0093] At processing block 614, processing logic initializes update attributes that were set to an update value. Updated mappings identify the pages that were modified by the guest OS.
  • processing logic synchronizes the shadow mappings based on modified guest pages and updates the metadata if needed due to the above modifications.
  • processing logic 618 determines whether a working set maintained by the VMM includes a shadow PD corresponding to the new guest PD requested by the VM. If so, processing logic requests the processor to load the base address of this shadow PT hierarchy (processing block 620).
  • FIG. 7 is a flow diagram of one embodiment of a process 700 for synchronizing entries of two translation data structures for a specified address.
  • Process 700 may be performed, for example, as a result of the INVLPG instruction issued by the VM or as a result of an induced page fault.
  • process 700 begins with processing logic determining whether the mapping in the shadow PT hierarchy for the specified address is stale (i.e., there is valid mapping that does not correspond to the current contents of the guest page table) (processing box 702). If not, processing logic proceeds to processing box 712. If so, processing logic determines whether the stale entry mapped a PD or PT page (processing box 704). If the stale entry did not map a PD or PT page, processing logic removes the stale entry (processing block 710) and proceeds to processing box 712.
  • processing logic determines whether the mapped page has been updated (processing box 706). If not, processing logic proceeds to processing block 710. If so, processing logic updates, synchronizes, or removes the modified PD or PT shadow(s) (processing block 708) and proceeds to processing block 710. In one embodiment, the page is marked for future synchronization.
  • processing logic determines whether the guest PT hierarchy contains a new mapping for the specified address. If not, process 700 ends. If so, processing logic adds the new mapping as a corresponding PTE or PDE and, if necessary, creates a shadow page and updates the metadata according to the addition (processing block 714).
  • Figure 8 is a flow diagram of one embodiment of a process 800 for removing a shadow PT hierarchy from a working set of shadow PT hierarchies maintained by the VMM.
  • a shadow PT hierarchy may be removed from the working set upon detecting a deactivation of a corresponding process by the VM.
  • the deactivation may be detected using heuristic defined for a relevant OS or employing a set of checks based on clues provided by the behavior of the guest VM with respect to the current address space. If the VM supports an interface through which the OS or a driver notifies the VMM of deactivations, then a heuristic may be avoided.
  • a shadow PT hierarchy may also be removed due to resource constraints, e.g., because the amount of memory used for shadow structures exceeds a target threshold.
  • processing logic begins with removing each valid PDE from the PD in the shadow PT hierarchy (processing block 802).
  • processing logic clears a corresponding entry in the PDV.
  • processing logic deallocates the PD page and removes the translation from a PD translation table (PDTT).
  • PDTT is used to track the address and type (e.g., PD or PT) of a page.
  • the PDTT is indexed by a guest PFN, with each entry containing a physical PFN and metadata.
  • processing logic removes monitoring from the PTEs that map the PD.
  • One embodiment of a process for removing monitoring from a PTE is discussed in more detail below in conjunction with Figure 14.
  • Figure 9 is a flow diagram of one embodiment of a process 900 for adding an entry to a PD of a shadow PT hierarchy. For illustration, we will consider a present entry which maps a page table.
  • processing logic begins with adding an entry for the PDE to the IPD (processing block 902).
  • processing logic determines if the PT mapped by this PDE is set in the PTV. If so, the appropriate shadow PT is looked up in the PTTT (processing block 916), the new shadow PDE is created (processing block 914) and process 900 ends.
  • processing logic sets a corresponding vector in the PTV (processing block 906), allocates a shadow page and initializes the translation (processing block 908), populates the new shadow page table (processing block 910), updates active PTE/PDE lists and metadata to reflect that the guest page used as a page table by the current guest PDE is to be monitored (processing block 912), and adds the new PDE, adding it to the active PDE list if the shadow page table contains any active PTE list elements (processing block 914).
  • One embodiment of a process for monitoring a PTE is discussed in more detail below in conjunction with Figure 13.
  • Figure 10 is a flow diagram of one embodiment of a process 1000 for removing an entry from a PD of a shadow PT hierarchy.
  • processing logic begins with removing an entry for this PDE from the IPD PDE list (processing block 1002). If the PDE is in the active PDE list, then the active PDE list must be updated.
  • processing logic determines whether the
  • PDE was the last entry to map the corresponding PT. If not, process 1000 ends. If so, processing logic clears the entry for the PT in the PTV (processing block 1006), removes each valid PTE (processing block 1008), updates the active PTE/PDE lists that map this PT for PT coverage (processing block 1010), and removes the shadow page translation and free the memory used to store the PT shadow page (processing block 1010).
  • Figure 11 is a flow diagram of one embodiment of a process 1100 for adding an entry to a PT of a shadow PT hierarchy.
  • processing logic begins with adding an entry for this PTE to the IPT (processing block 1102).
  • processing logic creates the shadow mapping and proceeds to processing box 1108.
  • processing logic determines whether a corresponding entry in the PDV or PTV is set. If not, process 1100 ends. If so, processing logic adds this entry to the active PTE list and updates associated metadata indicating if it maps a PD and/or PT page (processing block 1110). If the active PTE entry just created is the first for this page table, then the IPD must be consulted and each PDE which maps this page table page added to the active PDE list.
  • Figure 12 is a flow diagram of one embodiment of a process 1200 for removing an entry from a PT of a shadow PT hierarchy.
  • processing logic begins with determining whether this PTE maps a page set in the PDV or PTV (processing block 1202). If not, processing logic proceeds to processing block 1206. If so, processing logic removes the PTE from the active PTE list. If this was the last active PTE list element in the PT, then PDEs referencing this PT are removed from the active PDE list, (processing block 1204), and proceeds to processing block 1206.
  • processing logic removes the corresponding entry from the IPT.
  • Figure 13 is a flow diagram of one embodiment of a process 1300 for monitoring a PTE of a shadow PT hierarchy.
  • the steps shown in Figure 13 represent the processing that may be required when the monitor recognizes that a page which has been mapped as a data page is being used as a page directory or page table page. This process will therefore be triggered by a status change for the page mapped by the PTE.
  • processing logic begins with determining whether the PTE is identified in the active PTE list (processing box 1302). If so, processing logic adds the previously missing coverage (processing block 1304). This flow is triggered by a status change of the mapped page. Since this PTE was already in the active PTE list, it must be the case that this PTE was previously in use as a PT or PD, and is now in use in the other capacity as well. Such information may be explicitly stored with the entry or in associated metadata. If the PTE is not in the active PTE list, processing logic adds the PTE to the active PTE list and updates metadata accordingly (processing block 1306).
  • processing logic determines whether the PTE is the first active PTE list entry for this PT. If not, process 1300 ends. If so, processing logic adds, to the active PDE list, entries that map this PT (as found through the IPD) (processing block 1310).
  • Figure 14 is a flow diagram of one embodiment of a process 1400 for decreasing the monitoring coverage provided by a PTE of a shadow PT hierarchy.
  • This process might be invoked when a process is removed from the working set, or the last PDE to reference a page table is removed, resulting in a change of status of a previously monitored page directory or page table page.
  • processing logic begins with determining whether this PTE had monitored a page that was both a page table and page directory page (processing box 1402). If so, processing logic reduces the coverage level, indicating that the PTE now monitors a page as either a PT or as a PD, but not both (processing block 1404). If not, processing logic removes the PTE from the active PTE list (processing block 1406). Note that if the PTE was an element for a page tracked for its use in a single capacity, then it must now be the case that the page no longer requires monitoring.
  • processing logic removes the corresponding entries which mapped this page table from the active PDE list (as found through the IPD) (processing block 1410).
  • the physical or virtual platform may comprise multiple processors. Each processor may in turn comprise one or more threads or logical processors.
  • the processes discussed above can be used in a single-threaded system supporting a single-threaded VM or in a physical system with multiple logical processors that supports one or more VMs each containing a single virtual logical processor. Note that each VM has its own set of metadata, shadow page tables, etc. and that synchronization steps are confined to a given VM.

Abstract

In one embodiment, a method includes receiving control transitioned from a virtual machine (VM) due to a privileged event pertaining to a translation-lookaside buffer (TLB), and determining which entries in a guest translation data structure were modified by the VM. The determination is made based on metadata extracted from a shadow translation data structure maintained by a virtual machine monitor (VMM) and attributes associated with entries in the shadow translation data structure. The method further includes synchronizing entries in the shadow translation data structure that correspond to the modified entries in the guest translation data structure with the modified entries in the guest translation data structure.

Description

A METHOD AND APPARATUS FOR SUPPORTING ADDRESS TRANSLATION IN A VIRTUAL MACHINE ENVIRONMENT
Field [0001] Embodiments of the invention relate generally to virtual machines, and more specifically to supporting address translation in a virtual machine environment.
Background [0002] A conventional virtual-machine monitor (VMM) typically runs on a computer and presents to other software the abstraction of one or more virtual machines. Each virtual machine may function as a self-contained platform, running its own "guest operating system" (i.e., an operating system (OS) hosted by the VMM) and other software, collectively referred to as guest software. The guest software expects to operate as if it were running on a dedicated computer rather than a virtual machine. That is, the guest software expects to control various events and have access to hardware resources such as physical memory and memory-mapped input/output (I/O) devices. For example, the guest software expects to maintain control over address-translation operations and have the ability to allocate physical memory, provide protection from and between guest applications, use a variety of paging techniques, etc. However, in a virtual-machine environment, the VMM should be able to have ultimate control over the computer's resources to provide protection from and between virtual machines.
Brief Description of the Drawings
[0003] The present invention is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar elements and in which:
[0004] Figure 1 illustrates one embodiment of a virtual-machine environment, in which the present invention may operate;
[0005] Figure 2 illustrates operation of a virtual TLB, according to one embodiment of the present invention;
[0006] Figures 3A and 3B illustrate a process of creating and maintaining metadata for a shadow PT hierarchy, according to two alternative embodiments of the present invention; [0007] Figure 4 is a flow diagram of one embodiment of a process for synchronizing guest translation data structure and shadow translation data structure;
[0008] Figure 5 is a flow diagram of one embodiment of a process for maintaining metadata for a shadow translation data structure; [0009] Figure 6 is a flow diagram of one embodiment of a process for facilitating a change of an address space;
[0010] Figure 7 is a flow diagram of one embodiment of a process for synchronizing entries of two translation data structures for a specified address;
[0011] Figure 8 is a flow diagram of one embodiment of a process for removing a shadow PT hierarchy from a working set of shadow PT hierarchies maintained by the VMM;
[0012] Figure 9 is a flow diagram of one embodiment of a process for adding an entry to a PD of a shadow PT hierarchy;
[0013] Figure 10 is a flow diagram of one embodiment of a process for removing an entry from a PD of a shadow PT hierarchy;
[0014] Figure 11 is a flow diagram of one embodiment of a process for adding an entry to a PT of a shadow PT hierarchy;
[0015] Figure 12 is a flow diagram of one embodiment of a process for removing an entry from a PT of a shadow PT hierarchy; [0016] Figure 13 is a flow diagram of one embodiment of a process for monitoring a PTE of a shadow PT hierarchy; and
[0017] Figure 14 is a flow diagram of one embodiment of a process for removing monitoring from a PTE of a shadow PT hierarchy.
Description of Embodiments
[0018] A method and apparatus for supporting address translation in a virtual machine environment is described. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It will be apparent, however, to one skilled in the art that the present invention can be practiced without these specific details.
[0019] Some portions of the detailed descriptions that follow are presented in terms of algorithms and symbolic representations of operations on data bits within a computer system's registers or memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to convey most effectively the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of operations leading to a desired result. The operations are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like. [0020] It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from the following discussions, it is appreciated that throughout the present invention, discussions utilizing terms such as "processing" or "computing" or "calculating" or "determining" or the like, may refer to the action and processes of a computer system, or similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer-system memories or registers or other such information storage, transmission or display devices. [0021] In the following detailed description of the embodiments, reference is made to the accompanying drawings that show, by way of illustration, specific embodiments in which the invention may be practiced. In the drawings, like numerals describe substantially similar components throughout the several views. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. Other embodiments may be utilized and structural, logical, and electrical changes may be made without departing from the scope of the present invention. Moreover, it is to be understood that the various embodiments of the invention, although different, are not necessarily mutually exclusive. For example, a particular feature, structure, or characteristic described in one embodiment may be included within other embodiments. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims, along with the full scope of equivalents to which such claims are entitled. [0022] Although the below examples may describe providing support for address translation in a virtual machine environment in the context of execution units and logic circuits, other embodiments of the present invention can be accomplished by way of software. For example, in some embodiments, the present invention may be provided as a computer program product or software which may include a machine or computer- readable medium having stored thereon instructions which may be used to program a computer (or other electronic devices) to perform a process according to the present invention. In other embodiments, processes of the present invention might be performed by specific hardware components that contain hardwired logic for performing the processes, or by any combination of programmed computer components and custom hardware components.
[0023] Thus, a machine-readable medium may include any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer), but is not limited to, floppy diskettes, optical disks, Compact Disc, Read-Only Memories (CD- ROMs), and magneto-optical disks, Read-Only Memories (ROMs), Random Access Memories (RAMs), Erasable Programmable Read-Only Memories (EPROMs), Electrically Erasable Programmable Read-Only Memories (EEPROMs), magnetic or optical cards, flash memories, a transmission over the Internet, electrical, optical, acoustical or other forms of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc.) or the like.
[0024] Further, a design may go through various stages, from creation to simulation to fabrication. Data representing a design may represent the design in a number of manners. First, as is useful in simulations, the hardware may be represented using a hardware description language or another functional description language. Additionally, a circuit level model with logic and/or transistor gates may be produced at some stages of the design process. Furthermore, most designs, at some stage, reach a level of data representing the physical placement of various devices in the hardware model. In the case where conventional semiconductor fabrication techniques are used, data representing a hardware model may be the data specifying the presence or absence of various features on different mask layers for masks used to produce the integrated circuit. In any representation of the design, the data may be stored in any form of a machine- readable medium. An optical or electrical wave modulated or otherwise generated to transmit such information, a memory, or a magnetic or optical storage such as a disc may be the machine readable medium. Any of these mediums may "carry" or "indicate" the design or software information. When an electrical carrier wave indicating or carrying the code or design is transmitted, to the extent that copying, buffering, or re-transmission of the electrical signal is performed, a new copy is made. Thus, a communication provider or a network provider may make copies of an article (a carrier wave) embodying techniques of the present invention.
[0025] Figure 1 illustrates one embodiment of a virtual-machine environment 100, in which the present invention may operate. In this embodiment, bare platform hardware 116 comprises a computing platform, which may be capable, for example, of executing a standard operating system (OS) or a virtual-machine monitor (VMM), such as a VMM 112.
[0026] The VMM 112, typically implemented in software, may emulate and export a bare machine interface to higher level software. Such higher level software may comprise a standard or real-time OS, may be a highly stripped-down operating environment with limited operating system functionality, may not include traditional OS facilities, etc. Alternatively, for example, the VMM 112 may be run within, or on top of, another VMM. VMMs may be implemented, for example, in hardware, software, firmware or by a combination of various techniques.
[0027] The platform hardware 116 can be of a personal computer (PC), mainframe, handheld device, portable computer, set-top box, or any other computing system. The platform hardware 116 includes a processor 118 and memory 120.
[0028] Processor 118 can be any type of processor capable of executing software, such as a microprocessor, digital signal processor, microcontroller, or the like. The processor 118 may include microcode, programmable logic or hardcoded logic for performing the execution of method embodiments of the present invention. Although Figure 1 shows only one such processor 118, there may be one or more processors in the system.
[0029] Memory 120 can be a hard disk, a floppy disk, random access memory (RAM) (e.g., dynamic RAM (DRAM) or static RAM (SRAM)), read only memory (ROM), flash memory, any combination of the above devices, or any other type of machine medium readable by processor 118. Memory 120 may store instructions and/or data for performing the execution of method embodiments of the present invention. [0030] The VMM 112 presents to other software (i.e., "guest" software) the abstraction of one or more virtual machines (VMs), which may provide the same or different abstractions to the various guests. Figure 1 shows two VMs, 102 and 114. The guest software running on each VM may include a guest OS such as a guest OS 104 or 106 and various guest software applications 108 and 110. Each of the guest OSs 104 and 106 expects to access physical resources (e.g., processor registers, memory and I/O devices) within the VMs 102 and 114 on which the guest OS 104 or 106 is running and to perform other functions. For example, the guest OS 104 or 106 expects to have access to all registers, caches, structures, I/O devices, memory and the like, according to the architecture of the processor and platform presented in the VM 102 and 114. The resources that can be accessed by the guest software may either be classified as "privileged" or "non-privileged." For privileged resources, the VMM 112 facilitates functionality desired by guest software while retaining ultimate control over these privileged resources. Non-privileged resources do not need to be controlled by the VMM 112 and can be accessed directly by guest software.
[0031] Further, each guest OS expects to handle various fault events such as exceptions (e.g., page faults, general protection faults, etc.), interrupts (e.g., hardware interrupts, software interrupts), and platform events (e.g., initialization (INIT) and system management interrupts (SMIs)). Some of these fault events are "privileged" because they must be handled by the VMM 112 to ensure proper operation of VMs 102 and 114 and for protection from and among guest software.
[0032] When a privileged fault event occurs or guest software attempts to access a privileged resource, control may be transferred to the VMM 112. The transfer of control from guest software to the VMM 112 is referred to herein as a VM exit. After facilitating the resource access or handling the event appropriately, the VMM 112 may return control to guest software. The transfer of control from the VMM 112 to guest software is referred to as a VM entry.
[0033] In one embodiment, the processor 118 controls the operation of the VMs 102 and 114 in accordance with data stored in a virtual machine control structure (VMCS) 125. The VMCS 125 is a structure that may contain state of guest software, state of the VMM 112, execution control information indicating how the VMM 112 wishes to control operation of guest software, information controlling transitions between the VMM 112 and a VM, etc. The processor 118 reads information from the VMCS 125 to determine the execution environment of the VM and to constrain its behavior. In one embodiment, the ViVICS is stored in memory 120. In some embodiments, multiple VMCS structures are used to support multiple VMs.
[0034] During address translation operations, the VM 102 or 114 expects to allocate physical memory, provide protection from and between guest software applications (e.g., applications 108 or 110), use a variety of paging techniques, etc. In a non-virtual machine environment, an address translation mechanism expected by an OS may be based on a translation lookaside buffer (TLB) 122 controlled by the processor 118 and a translation data structure, such as a page-table (PT) hierarchy, controlled by the OS and used to translate virtual memory addresses into physical memory addresses when paging is enabled.
[0035] The architecture of the Intel® Pentium® 4 Processor supports a number of paging modes. The most commonly used paging mode supports a 32-bit linear address space using a two-level hierarchical paging structure (referred to herein as a two-level hierarchy paging mode). Embodiments of the invention are not limited to this paging mode, but instead may be employed by one skilled in the art to virtualize other paging modes (e.g., Physical Address Extension (PAE) mode, Intel® Extended Memory 64 Technology (EM64T) mode, etc.) and implementations (e.g., hashed page tables). In one embodiment based on a TLB, translation of a virtual memory address into a physical memory address begins with searching the TLB 122 using either the upper 20 bits (for a 4 KB page frame) or the upper 10 bits (for a 4MB page frame) of the virtual address. If a match is found (a TLB hit), the upper bits of a physical page frame that are contained in the TLB 122 are conjoined with the lower bits of the virtual address to form a physical address. The TLB also contains access and permission attributes associated with the mapping. If no match is found (a TLB miss), the processor consults the PT hierarchy to determine the virtual-to-physical translation, which is then cached in the TLB 122. Entries in the PT hierarchy may include some attributes that are automatically set by the processor on certain accesses.
[0036] If the PT hierarchy is modified, the TLB 122 may become inconsistent with the PT hierarchy if a corresponding address translation exists in the TLB 122. The OS may expect to be able to resolve such an inconsistency by issuing an instruction to the processor 118. For example, in the instruction set architecture (ISA) of the Intel® Pentium® 4 (referred to herein as the IA-32 ISA), a processor allows software to invalidate cached translations in the TLB by issuing the INVLPG instruction. In addition, the OS may expect to request the processor 118 to change the address space completely, which should result in the removal of all translations from the TLB 122. For example, in the IA- 32 ISA, an OS may use a MOV instruction or a task switch to request a processor to load CR3 (which contains the base address of the PT hierarchy), thereby removing all translations from the TLB. Different levels of the page table hierarchy may have different names based upon mode and implementation. In the two-level hierarchy paging mode, there are two levels of paging structures. The CR3 register points to the base of the page directory page. Entries in the page directory may either specify a mapping to a large-size page (e.g., a 4MB superpage, a 2MB superpage, IGB superpage, etc.), or a reference to a page table. The page table in turn may contain mappings to small-size pages.
[0037] As discussed above, in the virtual-machine environment, the VMM 112 should be able to have ultimate control over physical resources including the TLB 122. Embodiments of the present invention address the conflict between the expectations of the VMs 102 and 114 and the role of the VMM 112 by using a virtual TLB that emulates the functionality of the processor's physical TLB.
[0038] The virtual TLB includes the TLB 122 and a set of shadow PT hierarchies controlled by the VMM 112. The set of shadow PT hierarchies derive its format and content from guest PT hierarchies that may be currently used or not used by the VM 102 or 114. If the VM 102 or 114 modifies the content of the guest PT hierarchies, this content becomes inconsistent with the content of the shadow PT hierarchies. The inconsistencies between the guest PT hierarchies and the shadow PT hierarchies are resolved using techniques analogous to those employed by the processor 118 in managing the TLB 122. Some of these techniques force the VM 102 or 114 to issue an event indicating an attempt to manipulate the TLB (e.g., INVLPG, page fault, and load CR3). Such events are privileged and, therefore, result in a VM exit to the VMM 112. The VMM then evaluates the event and synchronizes all maintained shadow PT hierarchies with the current guest state if needed. We will refer to the set of maintained shadow PT hierarchies as the working set. As multiple processes may use the same guest page table, it is possible for the same shadow PT to be a part of multiple guest PT hierarchies. The corresponding shadow PT will in turn be a member of multiple shadow PT hierarchies. [0039] Note that synchronization performed by the VMM may update shadow page table or page directory entries for a shadow PT hierarchy that is not currently in-use. Likewise synchronization may be required to guest pages that are not part of the in-use guest PT hierarchy.
[0040] In one embodiment, the VMM 112 includes an address translation module 126 that is responsible for creating and maintaining a working set of shadow PT hierarchies for each of the VM 102 and 114 in a virtual TLB (VTLB) data store 124. The working set of shadow PT hierarchies is maintained for corresponding active processes of the VM 102 or 114 (i.e., processes that are likely to be activated in the near future by the VM 102 or 114). With the IA32 ISA, the only explicitly defined guest hierarchy is that defined by the currently used paging structures. In practice there is a high deal of temporal locality for guest processes and their address spaces. The VMM may employ heuristics or explicit information to determine a set of active process.
[0041] When the VM 102 or 114 enables a guest PT hierarchy for one of the active processes of the VM 102 or 114, the address translation module 126 identifies a corresponding shadow PT hierarchy in the working set and requests the processor 118 to load its base address. When applicable, the address translation module 126 can then reuse previously computed mappings that are stored in the shadow PT hierarchies.
[0042] If the VM 102 or 114 activates a new process, the address translation module 126 derives a new shadow PT hierarchy from a corresponding guest PT hierarchy and adds it to the working set. Alternatively, if the VM 102 or 114 de-activates an existing process, the address translation module 126 removes information corresponding to the guest PT hierarchy from the working set.
[0043] In one embodiment, the address translation module 126 is responsible for extracting metadata from each new shadow PT hierarchy, storing the metadata in the VTLB data store 124, and updating the metadata when the shadow PT hierarchy is modified. In one embodiment, the metadata includes a PT vector (PTV), a PD vector (PDV), an active PTE list, and an active PDE list.
[0044] The PTV and PDV track the guest frames that are used as PTs and PDs. In one embodiment, this information is encoded in bit vectors. The PTV may be indexed by page frame number (PFN), with each entry bit being set if a corresponding PFN is a PT. The PDV may be indexed by a page frame number (PFN), with each entry bit being set if a corresponding PFN is a PD.
[0045] The active PTE list is a list of PT entries (PTEs) in the shadow PT hierarchy that point to frames holding PTs and PD. The active PDE list identifies PD entries (PDEs) in the shadow PT hierarchy that point to PTs containing PT entries identified in the active PTE list.
[0046] In one embodiment, active PDE and PTE lists contain additional metadata describing whether the mapping is to a PD or PT frame. [0047] One skilled in the art will understand that embodiments of this invention may use a variety of data structures which may be more or less space or time efficient than those described herein. One skilled in the art will also recognize the extension of tracking structures to support additional paging modes. For example, an EM64T paging mode maps a 64-bit virtual address to a physical address through a four-level hierarchical paging structure. The actual number of bits supported in the virtual or physical address spaces may be implementation dependent and may be less than 64 bits in a particular implementation. As will be discussed in more detail below, an EM64T implementation may require additions of a page-map level 4 (PML4) page vector and a page directory pointer (PDP) page vector to track the additional page tables used in the EM64T paging structure. Likewise, one skilled in the art will recognize that the active PTE list will be extended to include entries which map any page used within the paging structures (e.g., PML4 or PDP pages for EM64T).
[0048] In one embodiment, active PTE/PDE list metadata is maintained to track the number of PD and PT frames that are mapped through a page table. When the number of mappings per page is incremented from 0, then PDEs which map the PT must be added to the active PDE list, and when the number of mappings is decreased to zero, then PDEs that map this PT must be removed from the active PDE list.
[0049] In one embodiment, the address translation module 126 is responsible for synchronizing a current shadow PT hierarchy with a current guest PT hierarchy when such synchronization is needed. The address translation module 126 performs the synchronization by determining which entries in the guest PT hierarchy have recently been modified and then updating corresponding entries in the shadow PT hierarchy accordingly. The address translation module 126 determines which entries in the guest PT hierarchy have recently been modified based on the metadata extracted from the shadow PT hierarchy and attributes associated with the entries of the shadow PT hierarchy. In one embodiment, the attributes include access attributes associated with PD entries in the shadow PT hierarchy and update attributes associated with PT entries in the shadow PT hierarchy. [0050] Figure 2 illustrates operation of a virtual TLB 204, according to one embodiment of the present invention. Virtual TLB 204 includes a shadow translation data structure represented by a shadow PT hierarchy 206 and a physical TLB 208. The shadow PT hierarchy 206 derives its structure and content from a guest translation data structure represented by a guest PT hierarchy 202. In one embodiment, the VMM maintains a working set of shadow PT hierarchies for active processes of the VM.
[0051] In one embodiment, when the VM requests the processor to enable a different guest PT hierarchy (e.g., by issuing MOV to CR3 or task switch in the IA-32 ISA), control transitions to the VMM, which instructs the processor to load the base address 214 of a shadow PT hierarchy 206 corresponding to the requested guest PT hierarchy 202. In some embodiments, this shadow PT hierarchy 206 is synchronized with the guest PT hierarchy 202 using relevant metadata and attributes, as will be discussed in greater detail below.
[0052] In one embodiment, the virtual TLB maintains access and update attributes in the entries of the shadow PD and PTs. These attributes are also referred to as an accessed (A) bit and a dirty (D) bit. In one embodiment, when a page frame is accessed by guest software for the first time, the processor sets the accessed (A) attribute in the corresponding PT entry or PD entry in the shadow PT hierarchy 206. If guest software attempts to write a page frame, the processor sets the dirty (D) attribute in the corresponding shadow PT entry.
[0053] Guest software is allowed to freely modify the guest PT hierarchy 202 including changing virtual-to-physical mapping, permissions, etc. Accordingly, the shadow PT hierarchy 206 may not be always consistent with the guest PT hierarchy 202. When a problem arises from an inconsistency between the hierarchies 202 and 206, the guest OS, which treats the virtual TLB 204 as a physical TLB, attempts to change the virtual TLB 204 by requesting a processor to perform an operation defined by a relevant ISA. For example, in the IA-32 ISA, such operations include the INVLPG instruction, CR3 loads, paging activation (modification of CR0.PG), modification of global paging (toggling of the CR4.PGE bit), etc. The operations attempting to change the virtual TLB 204 are configured by the VMM as privileged (e.g., using corresponding execution controls stored in the VMCS), and, therefore, result in a VM exit to the VMM. The VMM then determines the cause of the VM exit and modifies the content of the shadow PT hierarchy 206 if necessary. For example, if the VM exit occurs due to a page fault that should be handled by the guest OS (e.g., a page fault caused by an access not permitted by the guest PT hierarchy 202), the page fault is injected to the guest OS for handling. Alternatively, if the VM exit occurs due to a page fault (or any other operations such as INVLPG) resulting from an inconsistency between the entries of the hierarchies 202 and 206, the VMM may need to remove stale entries, add new entries, or modify existing entries, as will be discussed in more detail below. Page faults caused by the guest PT hierarchy are referred to herein as 'real' page faults, and page faults that would not have occurred with direct usage of the guest page tables are referred to herein as 'induced' page faults. [0054] Figure 3 A illustrates a process of creating and maintaining metadata for a shadow PT hierarchy in a two-level hierarchy paging mode, according to one embodiment of the present invention.
[0055] Referring to Figure 3 A, a number of physical page frames identified by distinct letters (letters A through W) is illustrated. Some guest page frames may contain a PD (e.g., frame A). Other guest page frames may contain a PT (e.g., frames A, B, C, and L). A hierarchy 302 is a guest PT hierarchy.
[0056] Figure 3 A shows a shadow PT hierarchy 304 created based on a guest PT hierarchy 302. Each PD or PT in the guest PT hierarchy 302 includes a corresponding PD or PT in the shadow PT hierarchy 304. Note that in general a shadow page is not required for each page in the guest PT. Some embodiments may choose to restrict shadow pages according to usage statistics (e.g., only generate shadow pages for guest PT pages that have been used), or according to resource constraints (e.g., maintaining only a set of shadow pages based on available memory). Separate shadow tables are maintained for PD and PT tables derived from the same physical frame. For example, separate tables 330 and 332 are maintained for PD 306 and PT 308 that are derived from the same physical frame 314. The PD and PT entries in the shadow PT hierarchy 304 contain transformed mappings for the guest frames 314 through 324.
[0057] In the guest PT hierarchy 302, frames 316 and 318 are used as PTs 310 and 312, and frame 314 is used both as PD 306 and PT 308. This usage is illustrated as "PT" and "PD/PT" in the page frames 314 through 316 shown under the shadow PT hierarchy 304.
[0058] The shadow PT hierarchy 304 is associated with an active PTE list 342 and an active PDE list 344. In one embodiment, the active PTE list 342 identifies PT entries in the shadow PT hierarchy 304 that map PT and PD page frames from the guest PT hierarchy 302. In particular, the active PTE list 342 identifies entries in the PT 332 that map page frames 314 through 318. In one embodiment, the active PDE list 344 identifies PD entries in the shadow PT hierarchy that point to PTs with entries identified in the active PTE list 342. In particular, the active PDE list 344 includes entries in the PD 330 that point to the PT 332. The active PTE list 342 and the active PDE list 344 are components of the metadata of the shadow PT hierarchy 304.
[0059] The shadow PT hierarchy 304 is associated with a PT bit vector (PTV) 362 and a PD bit vector (PDV) 364. In one embodiment, the PTV 362 tracks the guest page frames that are used as PTs. In particular, the PTV 362 includes page frames 314 through 318 which are used as PTs in the guest PT hierarchy 302. In one embodiment, the PDV 364 tracks the guest page frames that are used as PDs. In particular, the PDV 364 includes page frame 314 that is used as PD in the guest PT hierarchy 302. In one embodiment, the PTV 362 and PDV 364 represent all shadow PT hierarchies in the working set and track the capacity in which shadow pages are employed in the working set (e.g., if a shadow page has not been allocated for a guest PT, then the PTV will not reflect the guest PT page, even if it appears in the guest paging structures).
[0060] In one embodiment, if the guest OS adds a new PT to the guest PT hierarchy 302, the VMM may detect this addition (e.g., on the next or subsequent VM exit related to TLB manipulation) and add a corresponding PT to the shadow PT hierarchy 304. For example, if a new PT 352 derived from a frame 319 is added to the guest PT hierarchy 302, with a mapping for a new frame 354, the VMM may add a corresponding PT 360 with transformed mappings to the shadow PT hierarchy 304 and update the metadata to reflect this change. In particular, the VMM adds an entry mapping frame 319 in the PT 332 to the active PTE list 342, and an entry pointing to the PT 360 in the PD 330 to the active PDE list 344. Also, the VMM adds frame 319 to PTV 362, which tracks guest frames (i.e., here frame 319) used as PTs.
[0061] Figure 3B illustrates a process of creating and maintaining metadata for a shadow PT hierarchy in the EM64T paging mode, according to one embodiment of the present invention.
[0062] Referring to Figure 3B, the base of the paging structure is a PML4 page (e.g., frame A). Each entry in the PML4 page may reference a PDP page (e.g., frames B and C). Each entry in the PDP page may reference a page directory (PD) page (e.g., frame D or E), each entry of which in turn may reference a page in a page table (PT) page (e.g., frame F, G, H or I).
[0063] Each PML4, PDP, PD or PT page may be 4KB in size. In order to support physical address spaces larger than 32 bits, the entry size may be increased relative to the 32-bit paging mode. Specifically, there may be 512 entries per page, requiring that 9 bits of the virtual address be used at each level to select the appropriate entry. This selector size may lead to a large page size of 2MB instead of 4MB as described previously.
[0064] In one embodiment, the creation of metadata in the EM64T paging mode includes the generation of several vectors, an active entry list, and several active directory lists. The vectors include a PML4V vector identifying frames used as PML4 pages, a
PDPV vector identifying frames used as PDP pages, a PDV vector identifying frames used as PD pages, and a PTV vector identifying frames used as PT pages. The active entry list is an active PTE list including all mappings which map a PML4, PDP, PD or PT page. The active directory lists include lists identifying higher level mapping structures referencing a lower level structure through which the guest page corresponding to a shadow structure can be accessed. In particular, the active directory lists consist of an active PDE list including those PDEs that reference a page containing active PTE list entries, an active PDPE list including active PDPE entries which reference a PD containing an active PDE list entry, and an active PML4E list including entries which map a PDP containing elements in the active PDPE list.
[0065] In one embodiment, the synchronization of the shadow page tables begins with checking each entry in the active PML4E list associated with the used shadow PT hierarchy. If the entry has been accessed, each element in the active PDPE list corresponding to the accessed PML4 entry is checked, and then the processing continues as previously described.
[0066] In an alternative embodiment, active lists are not maintained and/or processed for one or more of the upper levels of the hierarchy. For example, in a system in which only a single entry is populated in the uppermost paging structure, the use of an active list for each level of the hierarchy will cause this single entry to be always accessed, thereby allowing no reduction in the amount of processing required for lower levels in the hierarchy. To accommodate this usage model, the synchronization may instead begin by processing an active list lower in the hierarchy. For example, in one embodiment, active PDPE list elements may first be processed followed by active PDE list elements or active ■ PTE list elements associated with a used shadow PT hierarchy. In one embodiment, the initial layer processed on synchronization may be predetermined. In another embodiment, the initial layer to be processed may be determined by dynamic profiling of the guest's page table usage. [0067] Various other paging modes may be used with embodiments of the present invention. For example, IA-32 supports an additional paging mode in which a 32-bit virtual address is mapped to a larger physical address. In this additional mode of operation, the page table base register is configured to point to a PDP page which contains four elements. Entry sizes and behaviors in this additional mode of operations are similar to those described above for the 64-bit virtual address mode. As this additional mode does not make use of PML4 pages, the PML4V and active PML4E list are not required. [0068] Figure 4 is a flow diagram of one embodiment of a process 400 for synchronizing a guest translation data structure and a shadow translation data structure. The process may be performed by processing logic that may comprise hardware (e.g., dedicated logic, programmable logic, microcode, etc.), software (such as that run on a general purpose computer system or a dedicated machine), or a combination of both. In one embodiment, the process is performed by an address translation module 126 of Figure 1.
[0069] Referring to Figure 4, process 400 begins with processing logic receiving control transitioned from a VM due to an event pertaining to manipulation of the TLB (processing block 402). Examples of such events may include a request to change the current address space (e.g., CR3 load), a request to adjust inconsistent translations for a specified virtual address in the TLB (e.g., INVLPG), a page fault, etc.
[0070] At processing box 404, processing logic determines whether the event pertaining to the manipulation of the TLB should be handled by the VM. If so (e.g., the event is a page fault caused by a problematic mapping in a guest translation data structure), control is returned to the VM for handling the event (processing block 406). If not, processing logic determines whether the event is associated with a specified problematic address (processing box 408). [0071] If the event does not need to be handled by the VM, it may be associated with a specified problematic address. Examples of such an event may include an event caused by the INVLPG instruction that takes a problematic address as an operand, an event caused by an induced page fault (e.g., a page fault resulting from an inconsistency between the two translation data structures with respect to a specific mapping, a page fault caused by a need to virtualize A/D bits in the guest translation data structure, etc.), etc. If the event is associated with a specified problematic address, processing logic makes corrections in the shadow translation data structure for the specified address (e.g., removes a stale mapping for the specified address or adds a new mapping for the specified address) to conform to the guest translation data structure (processing block 410). One embodiment of a process for synchronizing entries of two translation data structures for a specified address is discussed in more detail below in conjunction with Figure 7.
[0072] If the event is not associated with any specific address (e.g., the event is caused by a request of the VM to change the address space, which flushes all TLB entries in IA32), processing logic determines which entries of the guest translation data structure have been modified (processing block 412). The determination is made using metadata extracted from the shadow translation data structure and attributes associated with the entries of the shadow translation data structure (processing block 412). The metadata includes vectors and active lists for various levels of the shadow translation data structure. A vector for a specific level of the shadow translation data structure identifies frames used as pages at this level of the guest translation data structure. The active lists include an active entry list and one or more active directory lists. The active entry list includes mappings that map pages used by the guest in forming the guest translation data structure. The active directory lists identify higher level mapping structures referencing a lower level structure through which a guest page corresponding to a shadowed paging structure can be accessed. As discussed above, in the two-level hierarchy paging mode, the metadata includes, in one embodiment, vectors PTV and PDV, an active entry list (a PTE list), and an active directory list (a PDE list). In the EM64T paging mode, the metadata includes, in one embodiment, vectors PTV, PDV, PDPV and PML4V, an active entry list (a PTE list), and active directory lists (an active PDE list, an active PDPE list and an active PML4E list).
[0073] One embodiment of a process for identifying recently modified entries of the guest translation data structure using metadata is discussed in more detail below in conjunction with Figure 6.
[0074] At processing block 414, processing logic synchronizes corresponding entries in the shadow translation data structure with the modified entries of the guest translation data structure. Accordingly, processing logic only needs to synchronize the entries that were modified, rather than re-populating the entire content of the shadow translation data structure.
[0075] In one embodiment extra storage is used to maintain some guest PD and/or PT contents as they were last synchronized. This permits the VMM to determine where modifications have been made without calculating or looking up additional relocation or permission information.
[0076] Note that certain modifications to the guest page tables do not require modifications to the shadow page tables. For example, if a guest PT contains a not present mapping which is subsequently modified, no change is required to the corresponding shadow PT.
[0077] Figures 5-14 illustrate various processes performed to support address translation in a virtual machine environment using the two-level hierarchy paging mode, according to different embodiments of the present invention. These processes may be performed by processing logic that may comprise hardware (e.g., dedicated logic, programmable logic, microcode, etc.), software (such as that run on a general purpose computer system or a dedicated machine), or a combination of both. In one embodiment, each of these processes is performed by an address translation module 126 of Figure 1.
[0078] Figure 5 is a flow diagram of one embodiment of a process 500 for maintaining metadata for a shadow translation data structure such as a shadow PT hierarchy.
[0079] Referring to Figure 5, process 500 begins with processing logic creating a shadow page for each PD or PT page from the guest PT hierarchy (processing block 502).
[0080] At processing block 504, processing logic tracks page frames used as PDs or PTs in the guest PT hierarchy. In one embodiment, processing logic sets an entry in the PDV if a corresponding PFN is a PD in the guest PT hierarchy. Similarly, processing logic sets an entry in the PTV if a corresponding PFN is a PT in the guest PT hierarchy.
[0081] At processing block 506, processing logic tracks mappings to any Dynamic Random Access Memory (DRAM) backed page (to identify pages that can potentially be PDs or PTs). In one embodiment, processing logic tracks mappings to DRAM based pages using an inverted page table (IPT) and an inverted page directory (IPD). The IPT is indexed by a PFN of a data page frame, with each entry containing a list of addresses of PTEs that map the data page frame. The IPD is indexed by a PFN of the page table, with each entry containing a list of addresses of PDEs that reference the PFN as a page table. [0082] In one embodiment, at processing block 508, processing logic identifies 4MB pages in the guest PT hierarchy and creates a page table in the shadow PT hierarchy for each 4MB page to avoid large page mappings and thereby reduce future synchronization time. Otherwise, an update of a 4MB page would cause the synchronization of every PD and PT page within the 4MB. In one embodiment, an inverted expansion table (IET) is used to track which PDEs in the guest PT hierarchy point to a 4MB page. The IET is indexed by a PFN and attribute bits, with every entry listing PDEs that point to the exploded 4MB page.
[0083] In an embodiment of the invention the IPD may be indexed by the address of the shadow PFN to minimize required address translation steps.
[0084] In IA32, memory type information (e.g., cacheability information) can be stored in PAT bits within the PDE/PTE that maps a page. This type information is not captured in a PDE that is a page-table pointer. Hence, if two 4MB pages were to map the same region with different PAT attributes, then separate page tables would be required to convey the correct PAT attributes. Using separate expansion tables for each set of attributes resolves this issue.
[0085] At processing block 510, processing logic identifies PTEs in the shadow PT hierarchy that map pages used as PD or PT in the guest PT hierarchy and creates an active PTE list. [0086] At processing block 512, processing logic identifies PDEs in the shadow
PT hierarchy that point to PTs with PTEs identified in the active PTE list and creates an active PDE list.
[0087] Subsequently, at processing block 514, if the guest OS modifies the structure of the guest PT hierarchy (e.g., adds or removes a PD or PT), processing logic changes the above active PTE and PDE lists accordingly.
[0088] Figure 6 is a flow diagram of one embodiment of a process 600 for facilitating a change of an address space. Note that in IA32 the same CR3 value may also be reloaded to force a flush of stale TLB mappings. Similar processing steps are taken for a change of CR3 or for a CR3 reload. [0089] Referring to Figure 6, process 600 begins with processing logic determining that a VM exit occurred due to a request of the VM to enable a different guest PT hierarchy (e.g., by issuing a CR3 load request) (processing block 602). [0090] In response, processing logic scans all active PDEs corresponding to the currently in-use shadow PT hierarchy identified in the active PDE list of the metadata to find which of these PDEs have been accessed (have an access attribute set to an access value) (processing block 604), and then initializes the access attributes of the accessed PDEs (processing block 606). In IA32, nόn-leaf paging tables do not support a dirty bit. If the accessed bit is clear, then no page within the 4MB region has been read or written, so any guest page table or page directory cannot have been modified. However, the accessed bit does not distinguish between reads and writes, so 4MB regions which have been accessed should be further processed even though it is possible that nothing has been modified. In architectures supporting a dirty bit for non-leaf page tables, the dirty bit is checked instead, and only regions which had been written to require further processing.
[0091] Next, for each accessed PDE, processing logic scans all shadow PTEs corresponding to the accessed active PDE in the active PTE list of the metadata to find which of these PTEs include mappings for an updated page (have an update attribute set to an update value) (processing block 608).
[0092] Further, for each updated page, processing logic compares PD/PT entries in the guest PT hierarchy with corresponding entries in the shadow PT hierarchy (processing block 610) and changes the corresponding entries of the shadow PT hierarchy to conform to the modified entries of the guest PT hierarchy (e.g., by removing from the shadow PT hierarchy a PTE/PDE absent in the guest PT hierarchy, by adding to the shadow PT hierarchy a new PTE/PDE recently added to the guest PT hierarchy, etc.) (processing block 612). Note that adding PDEs may require the allocation and initialization of additional shadow PTs. This in turn may require updates to the various metadata structures maintained by the address translation module 126. [0093] At processing block 614, processing logic initializes update attributes that were set to an update value. Updated mappings identify the pages that were modified by the guest OS.
[0094] At processing block 616, processing logic synchronizes the shadow mappings based on modified guest pages and updates the metadata if needed due to the above modifications.
[0095] At processing logic 618, processing logic determines whether a working set maintained by the VMM includes a shadow PD corresponding to the new guest PD requested by the VM. If so, processing logic requests the processor to load the base address of this shadow PT hierarchy (processing block 620). If not, processing logic allocates a new shadow PT hierarchy corresponding to the requested guest PT hierarchy (processing block 622), adds the PD of the new shadow PT hierarchy to the PDV (processing block 624), adds each valid PDE to the PD of the new shadow PT hierarchy (processing block 626), configures the active PDE and PTE lists to monitor the PTEs that map this PD for PD coverage (processing block 628), and then requests the processor to load the base address of this shadow PT hierarchy (processing block 620). One embodiment of a process for monitoring a PTE is discussed in more detail below in conjunction with Figure 13. [0096] Figure 7 is a flow diagram of one embodiment of a process 700 for synchronizing entries of two translation data structures for a specified address. Process 700 may be performed, for example, as a result of the INVLPG instruction issued by the VM or as a result of an induced page fault.
[0097] Referring to Figure 7, process 700 begins with processing logic determining whether the mapping in the shadow PT hierarchy for the specified address is stale (i.e., there is valid mapping that does not correspond to the current contents of the guest page table) (processing box 702). If not, processing logic proceeds to processing box 712. If so, processing logic determines whether the stale entry mapped a PD or PT page (processing box 704). If the stale entry did not map a PD or PT page, processing logic removes the stale entry (processing block 710) and proceeds to processing box 712.
[0098] If the stale entry did map a PD or PT page, processing logic further determines whether the mapped page has been updated (processing box 706). If not, processing logic proceeds to processing block 710. If so, processing logic updates, synchronizes, or removes the modified PD or PT shadow(s) (processing block 708) and proceeds to processing block 710. In one embodiment, the page is marked for future synchronization.
[0099] At processing box 712, processing logic determines whether the guest PT hierarchy contains a new mapping for the specified address. If not, process 700 ends. If so, processing logic adds the new mapping as a corresponding PTE or PDE and, if necessary, creates a shadow page and updates the metadata according to the addition (processing block 714). [00100] Figure 8 is a flow diagram of one embodiment of a process 800 for removing a shadow PT hierarchy from a working set of shadow PT hierarchies maintained by the VMM.
[00101] A shadow PT hierarchy may be removed from the working set upon detecting a deactivation of a corresponding process by the VM. The deactivation may be detected using heuristic defined for a relevant OS or employing a set of checks based on clues provided by the behavior of the guest VM with respect to the current address space. If the VM supports an interface through which the OS or a driver notifies the VMM of deactivations, then a heuristic may be avoided. A shadow PT hierarchy may also be removed due to resource constraints, e.g., because the amount of memory used for shadow structures exceeds a target threshold.
[00102] Referring to Figure 8, processing logic begins with removing each valid PDE from the PD in the shadow PT hierarchy (processing block 802).
[00103] At processing block 804, processing logic clears a corresponding entry in the PDV.
[00104] At processing block 806, processing logic deallocates the PD page and removes the translation from a PD translation table (PDTT). The PDTT is used to track the address and type (e.g., PD or PT) of a page. The PDTT is indexed by a guest PFN, with each entry containing a physical PFN and metadata. [00105] At processing block 808, processing logic removes monitoring from the PTEs that map the PD. One embodiment of a process for removing monitoring from a PTE is discussed in more detail below in conjunction with Figure 14.
[00106] Figure 9 is a flow diagram of one embodiment of a process 900 for adding an entry to a PD of a shadow PT hierarchy. For illustration, we will consider a present entry which maps a page table.
[00107] Referring to Figure 9, processing logic begins with adding an entry for the PDE to the IPD (processing block 902).
[00108] At processing box 904, processing logic determines if the PT mapped by this PDE is set in the PTV. If so, the appropriate shadow PT is looked up in the PTTT (processing block 916), the new shadow PDE is created (processing block 914) and process 900 ends. If not, processing logic sets a corresponding vector in the PTV (processing block 906), allocates a shadow page and initializes the translation (processing block 908), populates the new shadow page table (processing block 910), updates active PTE/PDE lists and metadata to reflect that the guest page used as a page table by the current guest PDE is to be monitored (processing block 912), and adds the new PDE, adding it to the active PDE list if the shadow page table contains any active PTE list elements (processing block 914). One embodiment of a process for monitoring a PTE is discussed in more detail below in conjunction with Figure 13.
[00109] Figure 10 is a flow diagram of one embodiment of a process 1000 for removing an entry from a PD of a shadow PT hierarchy.
[00110] Referring to Figure 10, processing logic begins with removing an entry for this PDE from the IPD PDE list (processing block 1002). If the PDE is in the active PDE list, then the active PDE list must be updated.
[00111] At processing box 1004, processing logic determines whether the
PDE was the last entry to map the corresponding PT. If not, process 1000 ends. If so, processing logic clears the entry for the PT in the PTV (processing block 1006), removes each valid PTE (processing block 1008), updates the active PTE/PDE lists that map this PT for PT coverage (processing block 1010), and removes the shadow page translation and free the memory used to store the PT shadow page (processing block 1010).
[00112] Figure 11 is a flow diagram of one embodiment of a process 1100 for adding an entry to a PT of a shadow PT hierarchy.
[00113] Referring to Figure 11, processing logic begins with adding an entry for this PTE to the IPT (processing block 1102).
[00114] At processing box 1106, processing logic creates the shadow mapping and proceeds to processing box 1108.
[00115] At processing box 1108, processing logic determines whether a corresponding entry in the PDV or PTV is set. If not, process 1100 ends. If so, processing logic adds this entry to the active PTE list and updates associated metadata indicating if it maps a PD and/or PT page (processing block 1110). If the active PTE entry just created is the first for this page table, then the IPD must be consulted and each PDE which maps this page table page added to the active PDE list.
[00116] Figure 12 is a flow diagram of one embodiment of a process 1200 for removing an entry from a PT of a shadow PT hierarchy.
[00117] Referring to Figure 12, processing logic begins with determining whether this PTE maps a page set in the PDV or PTV (processing block 1202). If not, processing logic proceeds to processing block 1206. If so, processing logic removes the PTE from the active PTE list. If this was the last active PTE list element in the PT, then PDEs referencing this PT are removed from the active PDE list, (processing block 1204), and proceeds to processing block 1206.
[00118] At processing block 1206, processing logic removes the corresponding entry from the IPT.
[00119] Figure 13 is a flow diagram of one embodiment of a process 1300 for monitoring a PTE of a shadow PT hierarchy. The steps shown in Figure 13 represent the processing that may be required when the monitor recognizes that a page which has been mapped as a data page is being used as a page directory or page table page. This process will therefore be triggered by a status change for the page mapped by the PTE.
[00120] Referring to Figure 13, processing logic begins with determining whether the PTE is identified in the active PTE list (processing box 1302). If so, processing logic adds the previously missing coverage (processing block 1304). This flow is triggered by a status change of the mapped page. Since this PTE was already in the active PTE list, it must be the case that this PTE was previously in use as a PT or PD, and is now in use in the other capacity as well. Such information may be explicitly stored with the entry or in associated metadata. If the PTE is not in the active PTE list, processing logic adds the PTE to the active PTE list and updates metadata accordingly (processing block 1306). [00121] Next, at processing box 1308, processing logic determines whether the PTE is the first active PTE list entry for this PT. If not, process 1300 ends. If so, processing logic adds, to the active PDE list, entries that map this PT (as found through the IPD) (processing block 1310).
[00122] Figure 14 is a flow diagram of one embodiment of a process 1400 for decreasing the monitoring coverage provided by a PTE of a shadow PT hierarchy.
This process might be invoked when a process is removed from the working set, or the last PDE to reference a page table is removed, resulting in a change of status of a previously monitored page directory or page table page.
[00123] Referring to Figure 14, processing logic begins with determining whether this PTE had monitored a page that was both a page table and page directory page (processing box 1402). If so, processing logic reduces the coverage level, indicating that the PTE now monitors a page as either a PT or as a PD, but not both (processing block 1404). If not, processing logic removes the PTE from the active PTE list (processing block 1406). Note that if the PTE was an element for a page tracked for its use in a single capacity, then it must now be the case that the page no longer requires monitoring.
[00124] Next, if the last active PTE list element in the PT was removed
(processing box 1408), processing logic removes the corresponding entries which mapped this page table from the active PDE list (as found through the IPD) (processing block 1410).
[00125] As discussed above, the physical or virtual platform may comprise multiple processors. Each processor may in turn comprise one or more threads or logical processors. The processes discussed above can be used in a single-threaded system supporting a single-threaded VM or in a physical system with multiple logical processors that supports one or more VMs each containing a single virtual logical processor. Note that each VM has its own set of metadata, shadow page tables, etc. and that synchronization steps are confined to a given VM.
[00126] Thus, a method and apparatus for supporting address translation in a virtual machine environment have been described. It is to be understood that the above description is intended to be illustrative, and not restrictive. Many other embodiments will be apparent to those of skill in the art upon reading and understanding the above description. The scope of the invention should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims

CLAIMSWhat is claimed is:
1. A method comprising: receiving control transitioned from a virtual machine (VM) due to a privileged event pertaining to a translation-lookaside buffer (TLB); determining which entries in a guest translation data structure were modified by the VM, based on metadata extracted from a shadow translation data structure maintained by a virtual machine monitor (VMM) and attributes associated with entries in the shadow translation data structure; and synchronizing entries in the shadow translation data structure that correspond to the modified entries in the guest translation data structure with the modified entries in the guest translation data structure.
2. The method of claim 1 wherein: the guest translation data structure is used by the VM for address translation operations; and content of the shadow translation data structure is used by a processor to cache address translations in the TLB.
3. The method of claim 1 further comprising: maintaining a working set of shadow translation data structures, each shadow translation data structure in the working set being associated with one of a plurality of active processes of the VM; and reusing content of a shadow translation data structure from the working set that is associated with one of the plurality of active processes when a guest translation data structure associated with the one of the plurality of active processes is enabled.
4. The method of claim 3 wherein the privileged event is any one of a request of the VM to enable a different guest translation data structure, a page fault caused by one or more inconsistencies between entries of the guest translation data structure and entries of the shadow translation data structure, and a request of the VM to invalidate one or more address translations in a translation-lookaside buffer (TLB).
5. The method of claim 3 further comprising: determining that the privileged event is any one of a page fault, an INVLPG instruction, and a request to enable a new guest translation data structure; creating a new shadow translation data structure based on the new guest translation data structure; and deriving metadata from the new shadow translation data structure.
6. The method of claim 3 further comprising: determining that one of the plurality of the active processes of the VM is deactivated; and removing a shadow translation data structure associated with the active process being deactivated from the working set.
7. The method of claim 3 further comprising: determining one of the shadow translation data structures is no longer used; and removing the one of the shadow translation data structures from the working set.
8. The method of claim 1 wherein the metadata includes a set of vectors, an active entry list, and one or more active directory lists.
9. The method of claim 8 wherein: each vector in the set identifies frames used as pages at a corresponding level of the guest translation data structure; the active entry list identifies mappings that map pages used in forming the guest translation data structure for which a shadow translation data structure exists; and the one or more active directory lists identify higher level mapping structures referencing a lower level structure through which the shadow translation data structure can be accessed.
10. The method of claim 9 wherein: the active entry list is an active page table (PT) entry list, the active PT entry (PTE) list identifying PTEs in the shadow translation data structure that map PT pages and PD pages from the guest translation data structure, and the one or more active directory lists include an active page directory (PD) entry list, the active PD entry (PDE) list identifying PDEs in the shadow translation data structure that point to PTs having the PTEs from the active PTE list.
11. The method of claim 9 wherein: the active entry list is an active page table (PT) entry list, the active PT entry (PTE) list identifying mappings that map any of page map level 4 (PML4) pages, page directory pointer (PDP) pages, page directory (PD) pages, and PT pages; and the one or more active directory lists include an active PD entry (PDE) list containing PDEs that reference a page with active PTE list entries, an active PDP entry (PDPE) list containing active PDPE entries which reference a PD with an active PDE list entry, and an active PML4E entry (PML4E) list containing entries which map a PDP with elements from the active PDPE list.
12. The method of claim 10 wherein the attributes associated with entries in the shadow translation data structure include access attributes associated with PDEs in the shadow translation data structure and update attributes associated with PTEs in the shadow translation data structure.
13. The method of claim 12 wherein determining which entries in the guest translation data structure were modified by the VM comprises: identifying one or more PDEs from the active PDE list corresponding to the active PT hierarchy being synchronized that have access attributes set to an access value; and for each of the identified PDEs, finding corresponding PTEs from the active PTE list that have update attributes set to an update value.
14. The method of claim 13 further comprising: initializing the access attributes; and initializing the update attributes.
15. A method comprising: creating a shadow page table (PT) hierarchy based on a guest PT hierarchy used by a guest operating system for address translation operations; and deriving metadata from the shadow PT hierarchy to determine subsequently which entries of the guest PT hierarchy that are represented in the shadow PT hierarchy were modified, the metadata comprising an active entry list identifying mappings that map pages used by the guest operating system in forming the guest PT hierarchy, and one or more active directory lists identifying higher level mapping structures referencing a lower level structure through which the shadow PT hierarchy is to be accessed.
16. The method of claim 15 wherein: the active entry list is an active PT entry (PTE) list identifying PTEs in the shadow PT hierarchy, which map PT pages and page directory (PD) pages from the guest PT hierarchy, and the one or more active directory lists include an active PD entry (PDE) list identifying PDEs in the shadow PT hierarchy, which point to PTs with the identified PTEs.
17. The method of claim 15 wherein: the active entry list is an active PT entry (PTE) list identifying PTEs in the shadow PT hierarchy, which map any of page map level 4 (PML4) pages, page directory pointer (PDP) pages, page directory (PD) pages, and PT pages; and the one or more active directory lists include an active PD entry (PDE) list containing PDEs that reference a page with active PTE list entries, an active PDP entry (PDPE) list containing active PDPE entries which reference a PD with an active PDE list entry, and an active PML4E entry (PML4E) list containing entries which map a PDP with elements from the active PDPE list.
18. The method of claim 16 further comprising: tracking pages used by the guest operating system in forming the guest PT hierarchy; and tracking mappings to any Dynamic Random Access Memory (DRAM) backed page frame.
19. The method of claim 15 further comprising: identifying one or more large-size pages in the guest PT hierarchy; and creating a PT table in the shadow PT hierarchy for each of the one or more large- size pages.
20. The method of claim 15 further comprising: detecting a change in a structure of the guest PT hierarchy; and modifying the metadata to reflect the change.
21. The method of claim 15 wherein: content of the shadow PT hierarchy is used by a processor to cache address translations in the TLB.
22. An apparatus comprising: a guest translation data structure used by a virtual machine (VM) for address translation operations; a shadow translation data structure maintained by a virtual machine monitor
(VMM), the shadow translation data structure deriving a format and structure from the guest translation data structure; and an address translation module to determine, based on metadata extracted from the shadow translation data structure, which entries in the guest translation data structure were modified by the VM, and to synchronize entries in the shadow translation data structure that correspond to the modified entries in the guest translation data structure with the modified entries in the guest translation data structure.
23. The apparatus of claim 22 wherein the address translation module is further to maintain a working set of shadow translation data structures, each shadow translation data structure in the working set being associated with one of a plurality of active processes of the VM, and to reuse content of a shadow translation data structure from the working set that is associated with one of the plurality of active processes when a guest translation data structure associated with the one of the plurality of active processes is enabled.
24. The apparatus of claim 23 wherein the privileged event is any one of a request of the VM to enable a different guest translation data structure, a page fault caused by one or more inconsistencies between entries of the guest translation data structure and entries of the shadow translation data structure, and a request of the VM to invalidate one or more address translations in a translation-lookaside buffer (TLB).
25. A system comprising: a dynamic random access memory (DRAM) to store a guest translation data structure used by a virtual machine (VM) for address translation operations, and a shadow translation data structure deriving a format and structure from the guest translation data structure; and a processor, coupled to the DRAM, to determine, based on metadata extracted from the shadow translation data structure, which entries in the guest translation data structure were modified by the VM, and to synchronize entries in the shadow translation data structure that correspond to the modified entries in the guest translation data structure with the modified entries in the guest translation data structure.
26. The system of claim 25 wherein the processor is further to maintain a working set of shadow translation data structures, each shadow translation data structure in the working set being associated with one of a plurality of active processes of the VM, and to reuse content of a shadow translation data structure from the working set that is associated with one of the plurality of active processes when a guest translation data structure associated with the one of the plurality of active processes is enabled.
27. The system of claim 26 wherein the privileged event is any one of a request of the VM to enable a different guest translation data structure, a page fault caused by one or more inconsistencies between entries of the guest translation data structure and entries of the shadow translation data structure, and a request of the VM to invalidate one or more address translations in a translation-lookaside buffer (TLB).
28. A machine-readable medium containing instructions which, when executed by a processing system, cause the processing system to perform a method, the method comprising: receiving control transitioned from a virtual machine (VM) due to a privileged event pertaining to a translation-lookaside buffer (TLB); determining which entries in a guest translation data structure were modified by the VM, based on metadata extracted from a shadow translation data structure maintained by a virtual machine monitor (VMM) and attributes associated with entries in the shadow translation data structure; and synchronizing entries in the shadow translation data structure that correspond to the modified entries in the guest translation data structure with the modified entries in the guest translation data structure.
29. The machine-readable medium of claim 28 wherein the method further comprises: maintaining a working set of shadow translation data structures, each shadow translation data structure in the working set being associated with one of a plurality of active processes of the VM, and reusing content of a shadow translation data structure from the working set that is associated with one of the plurality of active processes when a guest translation data structure associated with the one of the plurality of active processes is enabled.
30. The machine-readable medium of claim 29 wherein the privileged event is any one of a request of the VM to enable a different guest translation data structure, a page fault caused by one or more inconsistencies between entries of the guest translation data structure and entries of the shadow translation data structure, and a request of the VM to invalidate one or more address translations in a translation-lookaside buffer (TLB).
PCT/US2006/003587 2005-01-28 2006-01-27 A method and apparatus for supporting address translation in a virtual machine environment WO2006081582A2 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
CN2006800031107A CN101107593B (en) 2005-01-28 2006-01-27 A method and apparatus for supporting address translation in a virtual machine environment
EP06720101.2A EP1856606B1 (en) 2005-01-28 2006-01-27 A method and apparatus for supporting address translation in a virtual machine environment
JP2007553378A JP5038907B2 (en) 2005-01-28 2006-01-27 Method and apparatus for supporting address translation in a virtual machine environment
KR1020077017491A KR100928353B1 (en) 2005-01-28 2006-01-27 Method and device for supporting address translation in virtual machine environment

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US11/045,524 US7395405B2 (en) 2005-01-28 2005-01-28 Method and apparatus for supporting address translation in a virtual machine environment
US11/045,524 2005-01-28

Publications (2)

Publication Number Publication Date
WO2006081582A2 true WO2006081582A2 (en) 2006-08-03
WO2006081582A3 WO2006081582A3 (en) 2007-04-12

Family

ID=36603334

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2006/003587 WO2006081582A2 (en) 2005-01-28 2006-01-27 A method and apparatus for supporting address translation in a virtual machine environment

Country Status (6)

Country Link
US (2) US7395405B2 (en)
EP (1) EP1856606B1 (en)
JP (1) JP5038907B2 (en)
KR (1) KR100928353B1 (en)
CN (1) CN101107593B (en)
WO (1) WO2006081582A2 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2441039A (en) * 2006-08-15 2008-02-20 Intel Corp Synchronising a Translation Lookaside Buffer to an Extended Paging Table
JP2008305106A (en) * 2007-06-06 2008-12-18 Hitachi Ltd Control program for virtual computer and virtual computer system
EP2278463A1 (en) * 2009-07-14 2011-01-26 Intel Corporation A method and apparatus for supporting address translation in a multiprocessor virtual machine environment
EP3640808A1 (en) * 2018-09-28 2020-04-22 Intel Corporation Split-control of page attributes between virtual machines and a virtual machine monitor

Families Citing this family (80)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7299337B2 (en) * 2005-05-12 2007-11-20 Traut Eric P Enhanced shadow page table algorithms
US7363463B2 (en) * 2005-05-13 2008-04-22 Microsoft Corporation Method and system for caching address translations from multiple address spaces in virtual machines
US8423747B2 (en) * 2008-06-30 2013-04-16 Intel Corporation Copy equivalent protection using secure page flipping for software components within an execution environment
US9785485B2 (en) * 2005-07-27 2017-10-10 Intel Corporation Virtualization event processing in a layered virtualization architecture
US8909946B2 (en) 2005-11-15 2014-12-09 Microsoft Corporation Efficient power management of a system with virtual machines
US7475183B2 (en) * 2005-12-12 2009-01-06 Microsoft Corporation Large page optimizations in a virtual machine environment
US8694712B2 (en) * 2006-12-05 2014-04-08 Microsoft Corporation Reduction of operational costs of virtual TLBs
US8615643B2 (en) * 2006-12-05 2013-12-24 Microsoft Corporation Operational efficiency of virtual TLBs
US9098347B2 (en) 2006-12-21 2015-08-04 Vmware Implementation of virtual machine operations using storage system functionality
US9189265B2 (en) 2006-12-21 2015-11-17 Vmware, Inc. Storage architecture for virtual machines
US9354927B2 (en) * 2006-12-21 2016-05-31 Vmware, Inc. Securing virtual machine data
US7788464B2 (en) * 2006-12-22 2010-08-31 Microsoft Corporation Scalability of virtual TLBs for multi-processor virtual machines
US7685401B2 (en) 2006-12-27 2010-03-23 Intel Corporation Guest to host address translation for devices to access memory in a partitioned system
WO2009001153A1 (en) * 2007-06-28 2008-12-31 Nokia Corporation Memory protection unit in a virtual processing environment
US8078827B2 (en) * 2007-07-05 2011-12-13 International Business Machines Corporation Method and apparatus for caching of page translations for virtual machines
US8739156B2 (en) * 2007-07-24 2014-05-27 Red Hat Israel, Ltd. Method for securing the execution of virtual machines
US8032897B2 (en) * 2007-07-31 2011-10-04 Globalfoundries Inc. Placing virtual machine monitor (VMM) code in guest context to speed memory mapped input/output virtualization
US8515075B1 (en) 2008-01-31 2013-08-20 Mcafee, Inc. Method of and system for malicious software detection using critical address space protection
GB2460393B (en) * 2008-02-29 2012-03-28 Advanced Risc Mach Ltd A data processing apparatus and method for controlling access to secure memory by virtual machines executing on processing circuitry
US8296547B2 (en) * 2008-08-27 2012-10-23 International Business Machines Corporation Loading entries into a TLB in hardware via indirect TLB entries
US10802990B2 (en) * 2008-10-06 2020-10-13 International Business Machines Corporation Hardware based mandatory access control
US10255463B2 (en) 2008-11-17 2019-04-09 International Business Machines Corporation Secure computer architecture
US8301863B2 (en) * 2008-11-17 2012-10-30 International Business Machines Corporation Recursive logical partition real memory map
US8135937B2 (en) * 2008-11-17 2012-03-13 International Business Machines Corporation Logical partition memory
JP4769308B2 (en) * 2009-01-20 2011-09-07 株式会社東芝 Virtual computer management mechanism, virtual computer system having the management mechanism, and paging processing method in the system
CN101882113B (en) * 2009-05-05 2012-02-22 北京大学 Memory virtualization method based on guest operation system kernel code replacement
JP4961459B2 (en) * 2009-06-26 2012-06-27 株式会社日立製作所 Virtual computer system and control method in virtual computer system
US8234407B2 (en) * 2009-06-30 2012-07-31 Oracle America, Inc. Network use of virtual addresses without pinning or registration
US8195917B2 (en) * 2009-07-01 2012-06-05 Advanced Micro Devices, Inc. Extended page size using aggregated small pages
US8560758B2 (en) * 2009-08-24 2013-10-15 Red Hat Israel, Ltd. Mechanism for out-of-synch virtual machine memory management optimization
US20110061050A1 (en) * 2009-09-04 2011-03-10 Sahita Ravi L Methods and systems to provide platform extensions for trusted virtual machines
US8327059B2 (en) * 2009-09-30 2012-12-04 Vmware, Inc. System and method to enhance memory protection for programs in a virtual machine environment
US9069592B2 (en) * 2009-11-02 2015-06-30 International Business Machines Corporation Generic transport layer mechanism for firmware communication
US8347290B2 (en) * 2009-11-16 2013-01-01 Novell, Inc. Monitoring spin locks in virtual machines in a computing system environment
US8656397B2 (en) 2010-03-30 2014-02-18 Red Hat Israel, Ltd. Migrating groups of threads across NUMA nodes based on remote page access frequency
US8364915B2 (en) * 2010-06-23 2013-01-29 Intel Corporation Method, apparatus and system for generating access information from an LRU tracking list
US8838935B2 (en) * 2010-09-24 2014-09-16 Intel Corporation Apparatus, method, and system for implementing micro page tables
KR101671494B1 (en) 2010-10-08 2016-11-02 삼성전자주식회사 Multi Processor based on shared virtual memory and Method for generating address translation table
KR20120071554A (en) * 2010-12-23 2012-07-03 한국전자통신연구원 Address space switching method and apparatus for full virtualization
US9767039B2 (en) 2011-07-18 2017-09-19 Vmware, Inc. Increasing granularity of dirty bit information in hardware assisted memory management systems
US8694738B2 (en) 2011-10-11 2014-04-08 Mcafee, Inc. System and method for critical address space protection in a hypervisor environment
US8973144B2 (en) 2011-10-13 2015-03-03 Mcafee, Inc. System and method for kernel rootkit protection in a hypervisor environment
US9069586B2 (en) * 2011-10-13 2015-06-30 Mcafee, Inc. System and method for kernel rootkit protection in a hypervisor environment
US9804870B2 (en) 2011-10-28 2017-10-31 Intel Corporation Instruction-set support for invocation of VMM-configured services without VMM intervention
CN104081346B (en) * 2012-02-07 2018-02-27 英特尔公司 For being interrupted using between tracking data Processing for removing device to support the method and apparatus of the address conversion in multiprocessor virtual machine environment
US9348757B2 (en) 2012-10-08 2016-05-24 International Business Machines Corporation System supporting multiple partitions with differing translation formats
US9600419B2 (en) 2012-10-08 2017-03-21 International Business Machines Corporation Selectable address translation mechanisms
US9740624B2 (en) 2012-10-08 2017-08-22 International Business Machines Corporation Selectable address translation mechanisms within a partition
US9355032B2 (en) 2012-10-08 2016-05-31 International Business Machines Corporation Supporting multiple types of guests by a hypervisor
US9280488B2 (en) 2012-10-08 2016-03-08 International Business Machines Corporation Asymmetric co-existent address translation structure formats
US9355040B2 (en) 2012-10-08 2016-05-31 International Business Machines Corporation Adjunct component to provide full virtualization using paravirtualized hypervisors
US9075789B2 (en) 2012-12-11 2015-07-07 General Dynamics C4 Systems, Inc. Methods and apparatus for interleaving priorities of a plurality of virtual processors
US9223602B2 (en) * 2012-12-28 2015-12-29 Intel Corporation Processors, methods, and systems to enforce blacklisted paging structure indication values
US10114662B2 (en) 2013-02-26 2018-10-30 Red Hat Israel, Ltd. Updating processor topology information for virtual machines
US10061622B2 (en) 2013-02-26 2018-08-28 Red Hat Israel, Ltd. Updating memory topology information for virtual machines
WO2014174580A1 (en) * 2013-04-22 2014-10-30 富士通株式会社 Information processing device, method, and program
WO2015061731A1 (en) 2013-10-27 2015-04-30 Advanced Micro Devices, Inc. Input/output memory map unit and northbridge
US9772867B2 (en) * 2014-03-27 2017-09-26 International Business Machines Corporation Control area for managing multiple threads in a computer
US9696933B2 (en) 2014-08-15 2017-07-04 International Business Machines Corporation Virtual machine manager initiated page-in of kernel pages
US9389897B1 (en) 2014-12-18 2016-07-12 International Business Machines Corporation Exiting multiple threads of a simulation environment in a computer
KR101751629B1 (en) 2014-12-24 2017-06-27 인텔 코포레이션 Hybrid on-demand graphics translation table shadowing
US9805194B2 (en) 2015-03-27 2017-10-31 Intel Corporation Memory scanning methods and apparatus
US9892060B2 (en) 2015-12-02 2018-02-13 International Business Machines Corporation Identifying stale entries in address translation cache
US10963280B2 (en) 2016-02-03 2021-03-30 Advanced Micro Devices, Inc. Hypervisor post-write notification of control and debug register updates
US10180909B2 (en) 2016-07-18 2019-01-15 International Business Machines Corporation Host-based resetting of active use of guest page table indicators
US10223281B2 (en) 2016-07-18 2019-03-05 International Business Machines Corporation Increasing the scope of local purges of structures associated with address translation
US10176110B2 (en) 2016-07-18 2019-01-08 International Business Machines Corporation Marking storage keys to indicate memory used to back address translation structures
US10168902B2 (en) 2016-07-18 2019-01-01 International Business Machines Corporation Reducing purging of structures associated with address translation
US10282305B2 (en) 2016-07-18 2019-05-07 International Business Machines Corporation Selective purging of entries of structures associated with address translation in a virtualized environment
US10802986B2 (en) 2016-07-18 2020-10-13 International Business Machines Corporation Marking to indicate memory used to back address translation structures
US10248573B2 (en) 2016-07-18 2019-04-02 International Business Machines Corporation Managing memory used to back address translation structures
US10241924B2 (en) 2016-07-18 2019-03-26 International Business Machines Corporation Reducing over-purging of structures associated with address translation using an array of tags
US10162764B2 (en) 2016-07-18 2018-12-25 International Business Machines Corporation Marking page table/page status table entries to indicate memory used to back address translation structures
US10176006B2 (en) 2016-07-18 2019-01-08 International Business Machines Corporation Delaying purging of structures associated with address translation
US10176111B2 (en) 2016-07-18 2019-01-08 International Business Machines Corporation Host page management using active guest page table indicators
US10169243B2 (en) 2016-07-18 2019-01-01 International Business Machines Corporation Reducing over-purging of structures associated with address translation
US10241925B2 (en) 2017-02-15 2019-03-26 Ati Technologies Ulc Selecting a default page size in a variable page size TLB
US10282309B2 (en) 2017-02-24 2019-05-07 Advanced Micro Devices, Inc. Per-page control of physical address space distribution among memory modules
US10339068B2 (en) 2017-04-24 2019-07-02 Advanced Micro Devices, Inc. Fully virtualized TLBs
EP3688583A1 (en) * 2017-09-26 2020-08-05 INTEL Corporation Methods and apparatus to process commands from virtual machines

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020082824A1 (en) * 2000-12-27 2002-06-27 Gilbert Neiger Virtual translation lookaside buffer
US20040064668A1 (en) * 2002-09-26 2004-04-01 Todd Kjos Memory addressing for a virtual machine implementation on a computer processor supporting virtual hash-page-table searching

Family Cites Families (231)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4319223A (en) * 1965-07-01 1982-03-09 The Bendix Corporation Full code means for electromechanical decoder
US3699532A (en) 1970-04-21 1972-10-17 Singer Co Multiprogramming control for a data handling system
US3996449A (en) 1975-08-25 1976-12-07 International Business Machines Corporation Operating system authenticator
US4162536A (en) 1976-01-02 1979-07-24 Gould Inc., Modicon Div. Digital input/output system and method
US4037214A (en) 1976-04-30 1977-07-19 International Business Machines Corporation Key register controlled accessing system
US4247905A (en) 1977-08-26 1981-01-27 Sharp Kabushiki Kaisha Memory clear system
US4278837A (en) 1977-10-31 1981-07-14 Best Robert M Crypto microprocessor for executing enciphered programs
US4276594A (en) 1978-01-27 1981-06-30 Gould Inc. Modicon Division Digital computer with multi-processor capability utilizing intelligent composite memory and input/output modules and method for performing the same
US4207609A (en) 1978-05-08 1980-06-10 International Business Machines Corporation Method and means for path independent device reservation and reconnection in a multi-CPU and shared device access system
JPS5823570B2 (en) 1978-11-30 1983-05-16 国産電機株式会社 Liquid level detection device
JPS5576447A (en) 1978-12-01 1980-06-09 Fujitsu Ltd Address control system for software simulation
US4307447A (en) 1979-06-19 1981-12-22 Gould Inc. Programmable controller
US4319323A (en) 1980-04-04 1982-03-09 Digital Equipment Corporation Communications device for data processing system
US4419724A (en) 1980-04-14 1983-12-06 Sperry Corporation Main bus interface package
US4366537A (en) 1980-05-23 1982-12-28 International Business Machines Corp. Authorization mechanism for transfer of program control or data between different address spaces having different storage protect keys
US4403283A (en) 1980-07-28 1983-09-06 Ncr Corporation Extended memory system and method
DE3034581A1 (en) 1980-09-13 1982-04-22 Robert Bosch Gmbh, 7000 Stuttgart READ-OUT LOCK FOR ONE-CHIP MICROPROCESSORS
JPS58140862A (en) 1982-02-16 1983-08-20 Toshiba Corp Mutual exclusion system
US4521852A (en) 1982-06-30 1985-06-04 Texas Instruments Incorporated Data processing device formed on a single semiconductor substrate having secure memory
JPS59111561A (en) 1982-12-17 1984-06-27 Hitachi Ltd Access controlling system of composite processor system
US4759064A (en) 1985-10-07 1988-07-19 Chaum David L Blind unanticipated signature systems
US4975836A (en) * 1984-12-19 1990-12-04 Hitachi, Ltd. Virtual computer system
JPS61206057A (en) 1985-03-11 1986-09-12 Hitachi Ltd Address converting device
FR2592510B1 (en) 1985-12-31 1988-02-12 Bull Cp8 METHOD AND APPARATUS FOR CERTIFYING SERVICES OBTAINED USING A PORTABLE MEDIUM SUCH AS A MEMORY CARD
FR2601535B1 (en) * 1986-07-11 1988-10-21 Bull Cp8 METHOD FOR CERTIFYING THE AUTHENTICITY OF DATA EXCHANGED BETWEEN TWO DEVICES CONNECTED LOCALLY OR REMOTELY THROUGH A TRANSMISSION LINE
FR2601525B1 (en) 1986-07-11 1988-10-21 Bull Cp8 SECURITY DEVICE PROHIBITING THE OPERATION OF AN ELECTRONIC ASSEMBLY AFTER A FIRST SHUTDOWN OF ITS POWER SUPPLY
FR2601476B1 (en) * 1986-07-11 1988-10-21 Bull Cp8 METHOD FOR AUTHENTICATING EXTERNAL AUTHORIZATION DATA BY A PORTABLE OBJECT SUCH AS A MEMORY CARD
FR2618002B1 (en) * 1987-07-10 1991-07-05 Schlumberger Ind Sa METHOD AND SYSTEM FOR AUTHENTICATING ELECTRONIC MEMORY CARDS
US5007082A (en) 1988-08-03 1991-04-09 Kelly Services, Inc. Computer software encryption apparatus
US5079737A (en) * 1988-10-25 1992-01-07 United Technologies Corporation Memory management unit for the MIL-STD 1750 bus
US5434999A (en) 1988-11-09 1995-07-18 Bull Cp8 Safeguarded remote loading of service programs by authorizing loading in protected memory zones in a terminal
FR2640798B1 (en) 1988-12-20 1993-01-08 Bull Cp8 DATA PROCESSING DEVICE COMPRISING AN ELECTRICALLY ERASABLE AND REPROGRAMMABLE NON-VOLATILE MEMORY
JPH02171934A (en) 1988-12-26 1990-07-03 Hitachi Ltd Virtual machine system
JPH02208740A (en) 1989-02-09 1990-08-20 Fujitsu Ltd Virtual computer control system
US5781753A (en) 1989-02-24 1998-07-14 Advanced Micro Devices, Inc. Semi-autonomous RISC pipelines for overlapped execution of RISC-like instructions within the multiple superscalar execution units of a processor having distributed pipeline control for speculative and out-of-order execution of complex instructions
US5442645A (en) 1989-06-06 1995-08-15 Bull Cp8 Method for checking the integrity of a program or data, and apparatus for implementing this method
JP2590267B2 (en) 1989-06-30 1997-03-12 株式会社日立製作所 Display control method in virtual machine
US5022077A (en) * 1989-08-25 1991-06-04 International Business Machines Corp. Apparatus and method for preventing unauthorized access to BIOS in a personal computer system
JP2825550B2 (en) 1989-09-21 1998-11-18 株式会社日立製作所 Multiple virtual space address control method and computer system
CA2010591C (en) 1989-10-20 1999-01-26 Phillip M. Adams Kernels, description tables and device drivers
CA2027799A1 (en) 1989-11-03 1991-05-04 David A. Miller Method and apparatus for independently resetting processors and cache controllers in multiple processor systems
US5075842A (en) 1989-12-22 1991-12-24 Intel Corporation Disabling tag bit recognition and allowing privileged operations to occur in an object-oriented memory protection mechanism
EP0473913A3 (en) 1990-09-04 1992-12-16 International Business Machines Corporation Method and apparatus for providing a service pool of virtual machines for a plurality of vm users
US5108590A (en) 1990-09-12 1992-04-28 Disanto Dennis Water dispenser
US5230069A (en) 1990-10-02 1993-07-20 International Business Machines Corporation Apparatus and method for providing private and shared access to host address and data spaces by guest programs in a virtual machine computer system
US5317705A (en) 1990-10-24 1994-05-31 International Business Machines Corporation Apparatus and method for TLB purge reduction in a multi-level machine system
US5287363A (en) 1991-07-01 1994-02-15 Disk Technician Corporation System for locating and anticipating data storage media failures
US5437033A (en) 1990-11-16 1995-07-25 Hitachi, Ltd. System for recovery from a virtual machine monitor failure with a continuous guest dispatched to a nonguest mode
US5255379A (en) 1990-12-28 1993-10-19 Sun Microsystems, Inc. Method for automatically transitioning from V86 mode to protected mode in a computer system using an Intel 80386 or 80486 processor
US5453003A (en) 1991-01-09 1995-09-26 Pfefferle; William C. Catalytic method
US5446904A (en) * 1991-05-17 1995-08-29 Zenith Data Systems Corporation Suspend/resume capability for a protected mode microprocessor
JPH04348434A (en) 1991-05-27 1992-12-03 Hitachi Ltd Virtual computer system
US5522075A (en) * 1991-06-28 1996-05-28 Digital Equipment Corporation Protection ring extension for computers having distinct virtual machine monitor and virtual machine address spaces
US5319760A (en) * 1991-06-28 1994-06-07 Digital Equipment Corporation Translation buffer for virtual machines with address space match
US5455909A (en) 1991-07-05 1995-10-03 Chips And Technologies Inc. Microprocessor with operation capture facility
JPH06236284A (en) 1991-10-21 1994-08-23 Intel Corp Method for preservation and restoration of computer-system processing state and computer system
US5627987A (en) 1991-11-29 1997-05-06 Kabushiki Kaisha Toshiba Memory management and protection system for virtual memory in computer system
US5574936A (en) 1992-01-02 1996-11-12 Amdahl Corporation Access control mechanism controlling access to and logical purging of access register translation lookaside buffer (ALB) in a computer system
US5486529A (en) 1992-04-16 1996-01-23 Zeneca Limited Certain pyridyl ketones for treating diseases involving leukocyte elastase
US5421006A (en) * 1992-05-07 1995-05-30 Compaq Computer Corp. Method and apparatus for assessing integrity of computer system software
US5237616A (en) 1992-09-21 1993-08-17 International Business Machines Corporation Secure computer system having privileged and unprivileged memories
US5293424A (en) 1992-10-14 1994-03-08 Bull Hn Information Systems Inc. Secure memory card
US5796835A (en) 1992-10-27 1998-08-18 Bull Cp8 Method and system for writing information in a data carrier making it possible to later certify the originality of this information
EP0600112A1 (en) 1992-11-30 1994-06-08 Siemens Nixdorf Informationssysteme Aktiengesellschaft Data processing system with virtual memory addressing and memory access controlled by keys
JP2765411B2 (en) 1992-11-30 1998-06-18 株式会社日立製作所 Virtual computer system
US5668971A (en) 1992-12-01 1997-09-16 Compaq Computer Corporation Posted disk read operations performed by signalling a disk read complete to the system prior to completion of data transfer
EP0602867A1 (en) 1992-12-17 1994-06-22 NCR International, Inc. An apparatus for securing a system platform
JPH06187178A (en) * 1992-12-18 1994-07-08 Hitachi Ltd Input and output interruption control method for virtual computer system
US5483656A (en) 1993-01-14 1996-01-09 Apple Computer, Inc. System for managing power consumption of devices coupled to a common bus
US5469557A (en) * 1993-03-05 1995-11-21 Microchip Technology Incorporated Code protection in microcontroller with EEPROM fuses
FR2703800B1 (en) 1993-04-06 1995-05-24 Bull Cp8 Method for signing a computer file, and device for implementing it.
FR2704341B1 (en) 1993-04-22 1995-06-02 Bull Cp8 Device for protecting the keys of a smart card.
JPH06348867A (en) 1993-06-04 1994-12-22 Hitachi Ltd Microcomputer
FR2706210B1 (en) * 1993-06-08 1995-07-21 Bull Cp8 Method for authenticating a portable object by an offline terminal, portable object and corresponding terminal.
US5555385A (en) 1993-10-27 1996-09-10 International Business Machines Corporation Allocation of address spaces within virtual machine compute system
US5825880A (en) 1994-01-13 1998-10-20 Sudia; Frank W. Multi-step digital signature method and system
US5459869A (en) 1994-02-17 1995-10-17 Spilo; Michael L. Method for providing protected mode services for device drivers and other resident software
US5604805A (en) 1994-02-28 1997-02-18 Brands; Stefanus A. Privacy-protected transfer of electronic information
FR2717286B1 (en) 1994-03-09 1996-04-05 Bull Cp8 Method and device for authenticating a data medium intended to allow a transaction or access to a service or a place, and corresponding medium.
US5684881A (en) 1994-05-23 1997-11-04 Matsushita Electric Industrial Co., Ltd. Sound field and sound image control apparatus and method
US5539828A (en) 1994-05-31 1996-07-23 Intel Corporation Apparatus and method for providing secured communications
US5473692A (en) * 1994-09-07 1995-12-05 Intel Corporation Roving software license for a hardware agent
US5533123A (en) 1994-06-28 1996-07-02 National Semiconductor Corporation Programmable distributed personal security
US5978481A (en) 1994-08-16 1999-11-02 Intel Corporation Modem compatible method and apparatus for encrypting data that is transparent to software applications
JPH0883211A (en) 1994-09-12 1996-03-26 Mitsubishi Electric Corp Data processor
DE69534757T2 (en) 1994-09-15 2006-08-31 International Business Machines Corp. System and method for secure storage and distribution of data using digital signatures
US6058478A (en) * 1994-09-30 2000-05-02 Intel Corporation Apparatus and method for a vetted field upgrade
FR2725537B1 (en) 1994-10-11 1996-11-22 Bull Cp8 METHOD FOR LOADING A PROTECTED MEMORY AREA OF AN INFORMATION PROCESSING DEVICE AND ASSOCIATED DEVICE
US5903752A (en) * 1994-10-13 1999-05-11 Intel Corporation Method and apparatus for embedding a real-time multi-tasking kernel in a non-real-time operating system
US5606617A (en) 1994-10-14 1997-02-25 Brands; Stefanus A. Secret-key certificates
US5564040A (en) 1994-11-08 1996-10-08 International Business Machines Corporation Method and apparatus for providing a server function in a logically partitioned hardware machine
US6269392B1 (en) 1994-11-15 2001-07-31 Christian Cotichini Method and apparatus to monitor and locate an electronic device using a secured intelligent agent
US5560013A (en) 1994-12-06 1996-09-24 International Business Machines Corporation Method of using a target processor to execute programs of a source architecture that uses multiple address spaces
US5555414A (en) 1994-12-14 1996-09-10 International Business Machines Corporation Multiprocessing system including gating of host I/O and external enablement to guest enablement at polling intervals
US5615263A (en) 1995-01-06 1997-03-25 Vlsi Technology, Inc. Dual purpose security architecture with protected internal operating system
US5764969A (en) 1995-02-10 1998-06-09 International Business Machines Corporation Method and system for enhanced management operation utilizing intermixed user level and supervisory level instructions with partial concept synchronization
FR2731536B1 (en) 1995-03-10 1997-04-18 Schlumberger Ind Sa METHOD FOR SECURE INFORMATION RECORDING ON A PORTABLE MEDIUM
US5717903A (en) 1995-05-15 1998-02-10 Compaq Computer Corporation Method and appartus for emulating a peripheral device to allow device driver development before availability of the peripheral device
JP3451595B2 (en) 1995-06-07 2003-09-29 インターナショナル・ビジネス・マシーンズ・コーポレーション Microprocessor with architectural mode control capable of supporting extension to two distinct instruction set architectures
US5684948A (en) 1995-09-01 1997-11-04 National Semiconductor Corporation Memory management circuit which provides simulated privilege levels
US5633929A (en) 1995-09-15 1997-05-27 Rsa Data Security, Inc Cryptographic key escrow system having reduced vulnerability to harvesting attacks
US5737760A (en) 1995-10-06 1998-04-07 Motorola Inc. Microcontroller with security logic circuit which prevents reading of internal memory by external program
US6093213A (en) 1995-10-06 2000-07-25 Advanced Micro Devices, Inc. Flexible implementation of a system management mode (SMM) in a processor
JP3693721B2 (en) 1995-11-10 2005-09-07 Necエレクトロニクス株式会社 Microcomputer with built-in flash memory and test method thereof
IL116708A (en) 1996-01-08 2000-12-06 Smart Link Ltd Real-time task manager for a personal computer
WO1997025798A1 (en) 1996-01-11 1997-07-17 Mrj, Inc. System for controlling access and distribution of digital property
US5657445A (en) 1996-01-26 1997-08-12 Dell Usa, L.P. Apparatus and method for limiting access to mass storage devices in a computer system
US5835594A (en) * 1996-02-09 1998-11-10 Intel Corporation Methods and apparatus for preventing unauthorized write access to a protected non-volatile storage
US5978892A (en) 1996-05-03 1999-11-02 Digital Equipment Corporation Virtual memory allocation in a virtual address space having an inaccessible gap
US5809546A (en) 1996-05-23 1998-09-15 International Business Machines Corporation Method for managing I/O buffers in shared storage by structuring buffer table having entries including storage keys for controlling accesses to the buffers
US6205550B1 (en) * 1996-06-13 2001-03-20 Intel Corporation Tamper resistant methods and apparatus
US6175925B1 (en) * 1996-06-13 2001-01-16 Intel Corporation Tamper resistant player for scrambled contents
US6178509B1 (en) * 1996-06-13 2001-01-23 Intel Corporation Tamper resistant methods and apparatus
US5729760A (en) 1996-06-21 1998-03-17 Intel Corporation System for providing first type access to register if processor in first mode and second type access to register if processor not in first mode
US5944821A (en) 1996-07-11 1999-08-31 Compaq Computer Corporation Secure software registration and integrity assessment in a computer system
US6199152B1 (en) * 1996-08-22 2001-03-06 Transmeta Corporation Translated memory protection apparatus for an advanced microprocessor
US5740178A (en) 1996-08-29 1998-04-14 Lucent Technologies Inc. Software for controlling a reliable backup memory
US6055637A (en) * 1996-09-27 2000-04-25 Electronic Data Systems Corporation System and method for accessing enterprise-wide resources by presenting to the resource a temporary credential
US5844986A (en) 1996-09-30 1998-12-01 Intel Corporation Secure BIOS
US5937063A (en) 1996-09-30 1999-08-10 Intel Corporation Secure boot
US5935242A (en) 1996-10-28 1999-08-10 Sun Microsystems, Inc. Method and apparatus for initializing a device
JPH10134008A (en) 1996-11-05 1998-05-22 Mitsubishi Electric Corp Semiconductor device and computer system
US5852717A (en) 1996-11-20 1998-12-22 Shiva Corporation Performance optimizations for computer networks utilizing HTTP
DE19649292A1 (en) * 1996-11-28 1998-06-04 Deutsche Telekom Ag Access protection method for pay television
US5901225A (en) * 1996-12-05 1999-05-04 Advanced Micro Devices, Inc. System and method for performing software patches in embedded systems
US5757919A (en) 1996-12-12 1998-05-26 Intel Corporation Cryptographically protected paging subsystem
US5818939A (en) 1996-12-18 1998-10-06 Intel Corporation Optimized security functionality in an electronic system
US6412035B1 (en) 1997-02-03 2002-06-25 Real Time, Inc. Apparatus and method for decreasing the response times of interrupt service routines
US5953502A (en) 1997-02-13 1999-09-14 Helbig, Sr.; Walter A Method and apparatus for enhancing computer system security
JP4000654B2 (en) 1997-02-27 2007-10-31 セイコーエプソン株式会社 Semiconductor device and electronic equipment
US6272637B1 (en) 1997-04-14 2001-08-07 Dallas Semiconductor Corporation Systems and methods for protecting access to encrypted information
US6557104B2 (en) * 1997-05-02 2003-04-29 Phoenix Technologies Ltd. Method and apparatus for secure processing of cryptographic keys
US6044478A (en) * 1997-05-30 2000-03-28 National Semiconductor Corporation Cache with finely granular locked-down regions
US6075938A (en) * 1997-06-10 2000-06-13 The Board Of Trustees Of The Leland Stanford Junior University Virtual machine monitors for scalable multiprocessors
US5987557A (en) 1997-06-19 1999-11-16 Sun Microsystems, Inc. Method and apparatus for implementing hardware protection domains in a system with no memory management unit (MMU)
US6175924B1 (en) * 1997-06-20 2001-01-16 International Business Machines Corp. Method and apparatus for protecting application data in secure storage areas
US6035374A (en) * 1997-06-25 2000-03-07 Sun Microsystems, Inc. Method of executing coded instructions in a multiprocessor having shared execution resources including active, nap, and sleep states in accordance with cache miss latency
US6584565B1 (en) 1997-07-15 2003-06-24 Hewlett-Packard Development Company, L.P. Method and apparatus for long term verification of digital signatures
US6014745A (en) * 1997-07-17 2000-01-11 Silicon Systems Design Ltd. Protection for customer programs (EPROM)
US6212635B1 (en) * 1997-07-18 2001-04-03 David C. Reardon Network security system allowing access and modification to a security subsystem after initial installation when a master token is in place
US5978475A (en) 1997-07-18 1999-11-02 Counterpane Internet Security, Inc. Event auditing system
US5919257A (en) 1997-08-08 1999-07-06 Novell, Inc. Networked workstation intrusion detection system
DE19735948C1 (en) * 1997-08-19 1998-10-01 Siemens Nixdorf Inf Syst Method for improving controllability in data processing equipment with translation-look-aside-buffer (TLB)
US6282657B1 (en) 1997-09-16 2001-08-28 Safenet, Inc. Kernel mode protection
US5935247A (en) 1997-09-18 1999-08-10 Geneticware Co., Ltd. Computer system having a genetic code that cannot be directly accessed and a method of maintaining the same
US6148379A (en) 1997-09-19 2000-11-14 Silicon Graphics, Inc. System, method and computer program product for page sharing between fault-isolated cells in a distributed shared memory system
US6182089B1 (en) * 1997-09-23 2001-01-30 Silicon Graphics, Inc. Method, system and computer program product for dynamically allocating large memory pages of different sizes
US6061794A (en) * 1997-09-30 2000-05-09 Compaq Computer Corp. System and method for performing secure device communications in a peer-to-peer bus architecture
US6357004B1 (en) * 1997-09-30 2002-03-12 Intel Corporation System and method for ensuring integrity throughout post-processing
US5970147A (en) 1997-09-30 1999-10-19 Intel Corporation System and method for configuring and registering a cryptographic device
US6085296A (en) 1997-11-12 2000-07-04 Digital Equipment Corporation Sharing memory pages and page tables among computer processes
US6219787B1 (en) 1997-12-22 2001-04-17 Texas Instruments Incorporated Method and apparatus for extending security model to native code
US6378072B1 (en) * 1998-02-03 2002-04-23 Compaq Computer Corporation Cryptographic system
US6308270B1 (en) 1998-02-13 2001-10-23 Schlumberger Technologies, Inc. Validating and certifying execution of a software program with a smart card
US6108644A (en) 1998-02-19 2000-08-22 At&T Corp. System and method for electronic transactions
US6131166A (en) 1998-03-13 2000-10-10 Sun Microsystems, Inc. System and method for cross-platform application level power management
US6192455B1 (en) * 1998-03-30 2001-02-20 Intel Corporation Apparatus and method for preventing access to SMRAM space through AGP addressing
US6374286B1 (en) * 1998-04-06 2002-04-16 Rockwell Collins, Inc. Real time processor capable of concurrently running multiple independent JAVA machines
US6173417B1 (en) * 1998-04-30 2001-01-09 Intel Corporation Initializing and restarting operating systems
US6397242B1 (en) * 1998-05-15 2002-05-28 Vmware, Inc. Virtualization system including a virtual machine monitor for a computer with a segmented architecture
FR2778998B1 (en) 1998-05-20 2000-06-30 Schlumberger Ind Sa METHOD FOR AUTHENTICATING A PERSONAL CODE OF A USER OF AN INTEGRATED CIRCUIT CARD
EP0961193B1 (en) 1998-05-29 2010-09-01 Texas Instruments Incorporated Secure computing device
US6421702B1 (en) 1998-06-09 2002-07-16 Advanced Micro Devices, Inc. Interrupt driven isochronous task scheduler system
US6505279B1 (en) * 1998-08-14 2003-01-07 Silicon Storage Technology, Inc. Microcontroller system having security circuitry to selectively lock portions of a program memory address space
US6339815B1 (en) * 1998-08-14 2002-01-15 Silicon Storage Technology, Inc. Microcontroller system having allocation circuitry to selectively allocate and/or hide portions of a program memory address space
JP2000076139A (en) 1998-08-28 2000-03-14 Nippon Telegr & Teleph Corp <Ntt> Portable information storage medium
US6363485B1 (en) * 1998-09-09 2002-03-26 Entrust Technologies Limited Multi-factor biometric authenticating device and method
US6463535B1 (en) 1998-10-05 2002-10-08 Intel Corporation System and method for verifying the integrity and authorization of software before execution in a local platform
US6230248B1 (en) 1998-10-12 2001-05-08 Institute For The Development Of Emerging Architectures, L.L.C. Method and apparatus for pre-validating regions in a virtual addressing scheme
US7194092B1 (en) 1998-10-26 2007-03-20 Microsoft Corporation Key-based secure storage
US6327652B1 (en) 1998-10-26 2001-12-04 Microsoft Corporation Loading and identifying a digital rights management operating system
US6330670B1 (en) 1998-10-26 2001-12-11 Microsoft Corporation Digital rights management operating system
US6609199B1 (en) 1998-10-26 2003-08-19 Microsoft Corporation Method and apparatus for authenticating an open system application to a portable IC device
US6445797B1 (en) 1998-12-16 2002-09-03 Secure Choice Llc Method and system for performing secure electronic digital streaming
US6463537B1 (en) 1999-01-04 2002-10-08 Codex Technologies, Inc. Modified computer motherboard security and identification system
US6282650B1 (en) 1999-01-25 2001-08-28 Intel Corporation Secure public digital watermark
US6560627B1 (en) * 1999-01-28 2003-05-06 Cisco Technology, Inc. Mutual exclusion at the record level with priority inheritance for embedded systems using one semaphore
US7111290B1 (en) * 1999-01-28 2006-09-19 Ati International Srl Profiling program execution to identify frequently-executed portions and to assist binary translation
US6188257B1 (en) * 1999-02-01 2001-02-13 Vlsi Technology, Inc. Power-on-reset logic with secure power down capability
EP1030237A1 (en) 1999-02-15 2000-08-23 Hewlett-Packard Company Trusted hardware device in a computer
US6272533B1 (en) 1999-02-16 2001-08-07 Hendrik A. Browne Secure computer system and method of providing secure access to a computer system including a stand alone switch operable to inhibit data corruption on a storage device
US7225333B2 (en) * 1999-03-27 2007-05-29 Microsoft Corporation Secure processor architecture for use with a digital rights management (DRM) system on a computing device
US6615278B1 (en) 1999-03-29 2003-09-02 International Business Machines Corporation Cross-platform program, system, and method having a global registry object for mapping registry equivalent functions in an OS/2 operating system environment
US6684326B1 (en) * 1999-03-31 2004-01-27 International Business Machines Corporation Method and system for authenticated boot operations in a computer system of a networked computing environment
US6651171B1 (en) 1999-04-06 2003-11-18 Microsoft Corporation Secure execution of program code
US6389537B1 (en) * 1999-04-23 2002-05-14 Intel Corporation Platform and method for assuring integrity of trusted agent communications
US6275933B1 (en) 1999-04-30 2001-08-14 3Com Corporation Security system for a computerized apparatus
EP1056014A1 (en) 1999-05-28 2000-11-29 Hewlett-Packard Company System for providing a trustworthy user interface
EP1055989A1 (en) 1999-05-28 2000-11-29 Hewlett-Packard Company System for digitally signing a document
US6321314B1 (en) 1999-06-09 2001-11-20 Ati International S.R.L. Method and apparatus for restricting memory access
US6633981B1 (en) 1999-06-18 2003-10-14 Intel Corporation Electronic system and method for controlling access through user authentication
US6158546A (en) 1999-06-25 2000-12-12 Tenneco Automotive Inc. Straight through muffler with conically-ended output passage
US6301646B1 (en) 1999-07-30 2001-10-09 Curl Corporation Pointer verification system and method
US6529909B1 (en) * 1999-08-31 2003-03-04 Accenture Llp Method for translating an object attribute converter in an information services patterns environment
JP2001148344A (en) 1999-09-09 2001-05-29 Nikon Corp Aligner, method for controlling output of energy source, laser using the method and method for manufacturing device
EP1085396A1 (en) 1999-09-17 2001-03-21 Hewlett-Packard Company Operation of trusted state in computing platform
US6535988B1 (en) * 1999-09-29 2003-03-18 Intel Corporation System for detecting over-clocking uses a reference signal thereafter preventing over-clocking by reducing clock rate
US6374317B1 (en) * 1999-10-07 2002-04-16 Intel Corporation Method and apparatus for initializing a computer interface
US6292874B1 (en) 1999-10-19 2001-09-18 Advanced Technology Materials, Inc. Memory management method and apparatus for partitioning homogeneous memory and restricting access of installed applications to predetermined memory ranges
EP1269425A2 (en) 2000-02-25 2003-01-02 Identix Incorporated Secure transaction system
AU2001243365A1 (en) 2000-03-02 2001-09-12 Alarity Corporation System and method for process protection
JP3710671B2 (en) 2000-03-14 2005-10-26 シャープ株式会社 One-chip microcomputer, IC card using the same, and access control method for one-chip microcomputer
CA2341931C (en) 2000-03-24 2006-05-30 Contentguard Holdings, Inc. System and method for protection of digital works
US6507904B1 (en) * 2000-03-31 2003-01-14 Intel Corporation Executing isolated mode instructions in a secure system running in privilege rings
US6633963B1 (en) 2000-03-31 2003-10-14 Intel Corporation Controlling access to multiple memory zones in an isolated execution environment
US6678825B1 (en) * 2000-03-31 2004-01-13 Intel Corporation Controlling access to multiple isolated memories in an isolated execution environment
US6921939B2 (en) * 2000-07-20 2005-07-26 Fairchild Semiconductor Corporation Power MOSFET and method for forming same using a self-aligned body implant
GB0020416D0 (en) * 2000-08-18 2000-10-04 Hewlett Packard Co Trusted system
US6715057B1 (en) * 2000-08-31 2004-03-30 Hewlett-Packard Development Company, L.P. Efficient translation lookaside buffer miss processing in computer systems with a large range of page sizes
US6938164B1 (en) 2000-11-22 2005-08-30 Microsoft Corporation Method and system for allowing code to be securely initialized in a computer
US7035963B2 (en) * 2000-12-27 2006-04-25 Intel Corporation Method for resolving address space conflicts between a virtual machine monitor and a guest operating system
US7631160B2 (en) 2001-04-04 2009-12-08 Advanced Micro Devices, Inc. Method and apparatus for securing portions of memory
US6976136B2 (en) 2001-05-07 2005-12-13 National Semiconductor Corporation Flash memory protection scheme for secured shared BIOS implementation in personal computers with an embedded controller
US7676430B2 (en) 2001-05-09 2010-03-09 Lenovo (Singapore) Ptd. Ltd. System and method for installing a remote credit card authorization on a system with a TCPA complaint chipset
EP1271277A3 (en) 2001-06-26 2003-02-05 Redstrike B.V. Security system and software to prevent unauthorized use of a computing device
US6523245B2 (en) * 2001-07-10 2003-02-25 Great Dane Limited Partnership Automated drill and rivet machine
US20030018892A1 (en) * 2001-07-19 2003-01-23 Jose Tello Computer with a modified north bridge, security engine and smart card having a secure boot capability and method for secure booting a computer
US7191464B2 (en) * 2001-10-16 2007-03-13 Lenovo Pte. Ltd. Method and system for tracking a secure boot in a trusted computing environment
US6961806B1 (en) * 2001-12-10 2005-11-01 Vmware, Inc. System and method for detecting access to shared structures and for maintaining coherence of derived structures in virtualized multiprocessor systems
US7103771B2 (en) 2001-12-17 2006-09-05 Intel Corporation Connecting a virtual token to a physical token
US20030126453A1 (en) 2001-12-31 2003-07-03 Glew Andrew F. Processor supporting execution of an authenticated code instruction
US7308576B2 (en) 2001-12-31 2007-12-11 Intel Corporation Authenticated code module
US7107460B2 (en) 2002-02-15 2006-09-12 International Business Machines Corporation Method and system for securing enablement access to a data security device
US7343493B2 (en) 2002-03-28 2008-03-11 Lenovo (Singapore) Pte. Ltd. Encrypted file system using TCPA
US7136867B1 (en) * 2002-04-08 2006-11-14 Oracle International Corporation Metadata format for hierarchical data storage on a raw storage device
US7318141B2 (en) 2002-12-17 2008-01-08 Intel Corporation Methods and systems to control virtual machines
US7389308B2 (en) * 2003-05-30 2008-06-17 Microsoft Corporation Shadow paging
US7310721B2 (en) * 2003-10-30 2007-12-18 Microsoft Corporation Shadow page tables for address translation control
US7562179B2 (en) * 2004-07-30 2009-07-14 Intel Corporation Maintaining processor resources during architectural events
US7281102B1 (en) * 2004-08-12 2007-10-09 Vmware, Inc. Restricting memory access to protect data when sharing a common address space
US7363463B2 (en) * 2005-05-13 2008-04-22 Microsoft Corporation Method and system for caching address translations from multiple address spaces in virtual machines

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020082824A1 (en) * 2000-12-27 2002-06-27 Gilbert Neiger Virtual translation lookaside buffer
US20040064668A1 (en) * 2002-09-26 2004-04-01 Todd Kjos Memory addressing for a virtual machine implementation on a computer processor supporting virtual hash-page-table searching

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
BARHAM P ET AL: "Xen and the Art of Virtualization" PROCEEDINGS OF THE ACM SYMPOSIUM ON OPERATING SYSTEMS PRINCIPLES, XX, XX, October 2003 (2003-10), pages 164-177, XP002298786 *

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9251094B2 (en) 2006-08-15 2016-02-02 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US9678890B2 (en) 2006-08-15 2017-06-13 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
GB2441039A (en) * 2006-08-15 2008-02-20 Intel Corp Synchronising a Translation Lookaside Buffer to an Extended Paging Table
US7555628B2 (en) 2006-08-15 2009-06-30 Intel Corporation Synchronizing a translation lookaside buffer to an extended paging table
US9262338B1 (en) 2006-08-15 2016-02-16 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US8099581B2 (en) 2006-08-15 2012-01-17 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US8296546B2 (en) 2006-08-15 2012-10-23 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US9298641B2 (en) 2006-08-15 2016-03-29 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US8949571B2 (en) 2006-08-15 2015-02-03 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US20150205723A1 (en) * 2006-08-15 2015-07-23 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US9122624B2 (en) 2006-08-15 2015-09-01 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US9141555B2 (en) * 2006-08-15 2015-09-22 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
GB2441039B (en) * 2006-08-15 2009-01-28 Intel Corp Synchronizing a translation lookaside buffer to an extended paging table
DE102007063960B3 (en) 2006-08-15 2021-12-16 Intel Corporation Synchronizing a translation buffer (TLB) with an extended paging table
US8601233B2 (en) 2006-08-15 2013-12-03 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US9298640B2 (en) 2006-08-15 2016-03-29 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US9330021B2 (en) 2006-08-15 2016-05-03 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US9372807B2 (en) 2006-08-15 2016-06-21 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
US9372806B2 (en) 2006-08-15 2016-06-21 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
DE102007037814B4 (en) 2006-08-15 2021-12-02 Intel Corporation Synchronize a translation lookaside buffer with an extended paging table
US10747682B2 (en) 2006-08-15 2020-08-18 Intel Corporation Synchronizing a translation lookaside buffer with an extended paging table
JP2008305106A (en) * 2007-06-06 2008-12-18 Hitachi Ltd Control program for virtual computer and virtual computer system
EP2278463A1 (en) * 2009-07-14 2011-01-26 Intel Corporation A method and apparatus for supporting address translation in a multiprocessor virtual machine environment
EP3640808A1 (en) * 2018-09-28 2020-04-22 Intel Corporation Split-control of page attributes between virtual machines and a virtual machine monitor
US10768968B2 (en) 2018-09-28 2020-09-08 Intel Corporation Split-control of page attributes between virtual machines and a virtual machine monitor

Also Published As

Publication number Publication date
WO2006081582A3 (en) 2007-04-12
CN101107593A (en) 2008-01-16
US20060174053A1 (en) 2006-08-03
US20090006805A1 (en) 2009-01-01
EP1856606B1 (en) 2016-05-11
US7836275B2 (en) 2010-11-16
KR100928353B1 (en) 2009-11-23
JP5038907B2 (en) 2012-10-03
EP1856606A2 (en) 2007-11-21
KR20070090047A (en) 2007-09-04
US7395405B2 (en) 2008-07-01
CN101107593B (en) 2013-02-06
JP2008529176A (en) 2008-07-31

Similar Documents

Publication Publication Date Title
EP1856606B1 (en) A method and apparatus for supporting address translation in a virtual machine environment
US9405567B2 (en) Method and apparatus for supporting address translation in a multiprocessor virtual machine environment using tracking data to eliminate interprocessor interrupts
US20110016290A1 (en) Method and Apparatus for Supporting Address Translation in a Multiprocessor Virtual Machine Environment
US7363463B2 (en) Method and system for caching address translations from multiple address spaces in virtual machines
US10324863B2 (en) Protected memory view for nested page table access by virtual machine guests
US8661181B2 (en) Memory protection unit in a virtual processing environment
US8762684B2 (en) Hardware assistance for page table coherence with guest page mappings
US8443156B2 (en) Virtualization system using hardware assistance for shadow page table coherence
US8316211B2 (en) Generating multiple address space identifiers per virtual machine to switch between protected micro-contexts
US20140108701A1 (en) Memory protection unit in a virtual processing environment
US20080005447A1 (en) Dynamic mapping of guest addresses by a virtual machine monitor
US9684605B2 (en) Translation lookaside buffer for guest physical addresses in a virtual machine
US9251102B2 (en) Virtualizing processor memory protection with “L1 iterate and L2 drop/repopulate”
US8621136B2 (en) Virtualizing processor memory protection with “L1 iterate and L2 swizzle”

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200680003110.7

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 2007553378

Country of ref document: JP

WWE Wipo information: entry into national phase

Ref document number: 1020077017491

Country of ref document: KR

WWE Wipo information: entry into national phase

Ref document number: 2006720101

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: CR2008-009891

Country of ref document: CR