摘要:
The present invention is directed to making a guest operating system aware of the topology of the subset of host resources currently assigned to it. At virtual machine boot time a Static Resource Affinity Table (SRAT) will be used by the virtualizer to group guest physical memory and guest virtual processors into virtual nodes. Thereafter, in one embodiment, the host physical memory behind a virtual node can be changed by the virtualizer as necessary, and the virtualizer will provide physical processors appropriate for the virtual processors in that node.
摘要:
The present invention is a system for and method of initializing multiple virtual processors in a virtual machine (VM) environment. The method of initializing multiple virtual processors includes the steps of the host creating a multiple processor VM and activating a “starter virtual processor,” the “starter virtual processor” issuing a startup command to a next virtual processor, the virtual machine monitor (VMM) giving the target virtual processor the highest priority for accessing the hardware resources, the VMM forcing the “starter virtual processor” to relinquish control of the hardware resources, the VMM handing control of the hardware resources to the target virtual processor, the target virtual processor executing and completing its startup routine, the VMM forcing the target virtual processor to relinquish control of the hardware resources, and the VMM handing control of the hardware resources back to the “starter virtual processor” for activating subsequent virtual processors.
摘要:
Techniques for reducing virtual machine input/output emulation overhead and decreasing the attack surface of a virtual machine architecture are disclosed.
摘要:
Various embodiments of the present invention are directed to augmented interrupt controllers (AICs) and to synthetic interrupt sources (SISs) providing richer interrupt information (or “synthetic interrupts” or “SIs”). The AIC and SIS provide efficient means for sending and receiving interrupts, and particularly interrupts sent to and received by virtual machines. Several of these embodiments are specifically directed to an interrupt controller that is extended to accept and deliver additional information associated with an incoming interrupt. For certain such embodiments, a memory-mapped extension to the interrupt controller includes a data structure that is populated with the additional information as part of the interrupt delivery. Although several of the embodiments described herein are disclosed in the context of a virtual machine system, the inventions disclosed herein can also be applied to traditional computer systems (without a virtualization layer) as well.
摘要:
Various embodiments of the present invention are directed to a multi-level virtualizer that is designed to remove much of the intercept-related functionality from the base-level virtualizer (that exists outside of each partition) and, instead, incorporate much of this functionality directly into each partition. For several of the embodiments, certain intercept handling functions are performed by an “external monitor” that runs within a partition and responds to specific intercept events, and the base-level virtualizer installs these external monitors within each partition and thereafter manages the external monitors for both single-partition and cross-partition intercept events. This distributed approach to intercept handling allows for a much less complex virtualizer and moves the intercept functionality up into each partition where each external monitor uses the resources of the corresponding guest operating system in that partition to resolve the intercept event.
摘要:
Enhanced shadow page table algorithms are presented for enhancing typical page table algorithms. In a virtual machine environment, where an operating system may be running within a partition, the operating system maintains it's own guest page tables. These page tables are not the real page tables that map to the real physical memory. Instead, the memory is mapped by shadow page tables maintained by a virtualing program, such as a hypervisor, that virtualizes the partition containing the operating system. Enhanced shadow page table algorithms provide efficient ways to harmonize the shadow page tables and the guest page tables. Specifically, by using tagged translation lookaside buffers, batched shadow page table population, lazy flags, and cross-processor shoot downs, the algorithms make sure that changes in the guest pages tables are reflected in the shadow page tables.
摘要:
Various mechanisms are disclosed for improving the operational efficiency of a virtual translation look-aside buffer (TLB) in a virtual machine environment. For example, one mechanism fills in entries in a shadow page table (SPT) and additionally, speculatively fills in other entries in the SPT based on various heuristics. Another mechanism allows virtual TLBs (translation look-aside buffers) to cache partial walks in a guest page table tree. Still another mechanism allows for dynamic resizing of the virtual TLB to optimize for run-time characteristics of active workloads. Still another mechanism allows virtual machine monitors (VMMs) to support legacy and enlightened modes of virtual TLB operation. Finally, another mechanism allows the VMM to remove only the stale entries in SPTs when linking or switching address spaces. All these mechanisms, together or in part, increase the operational efficiency of the virtual TLB.
摘要:
The present invention is a system and method that performs disk migration in a virtual machine environment. The present invention quickly and easily migrates a virtual machine from one host to another thus improving flexibility and efficiency in a virtual machine environment for “load balancing” systems, performing hardware or software upgrades, handling disaster recovery, and so on. Certain of the embodiments are specifically directed to providing a mechanism for migrating the disk state along with the device and memory states, where the disk data resides in a remotely located storage device that is common to multiple host computer systems in a virtual machine environment. The virtual machine migration process, which includes disk data migration, occurs without the user's awareness and, therefore, without the user's experiencing any noticeable interruption.
摘要:
A method of virtualizing memory through shadow page tables that cache translations from multiple guest address spaces in a virtual machine includes a software version of a hardware tagged translation look-aside buffer. Edits to guest page tables are detected by intercepting the creation of guest-writable mappings to guest page tables with translations cached in shadow page tables. The affected cached translations are marked as stale and purged upon an address space switch or an indiscriminate flush of translations by the guest. Thereby, non-stale translations remain cached but stale translations are discarded. The method includes tracking the guest-writable mappings to guest page tables, deferring discovery of such mappings to a guest page table for the first time until a purge of all cached translations when the number of untracked guest page tables exceeds a threshold, and sharing shadow page tables between shadow address spaces and between virtual processors.
摘要:
Enhanced shadow page table algorithms are presented for enhancing typical page table algorithms. In a virtual machine environment, where an operating system may be running within a partition, the operating system maintains it's own guest page tables. These page tables are not the real page tables that map to the real physical memory. Instead, the memory is mapped by shadow page tables maintained by a virtualing program, such as a hypervisor, that virtualizes the partition containing the operating system. Enhanced shadow page table algorithms provide efficient ways to harmonize the shadow page tables and the guest page tables. Specifically, by using tagged translation lookaside buffers, batched shadow page table population, lazy flags, and cross-processor shoot downs, the algorithms make sure that changes in the guest pages tables are reflected in the shadow page tables.