Microsoft Technology Licensing, LLC

Google Inc.

Andrew John Thornton

Jacob Kappeler Oshins

Marcus Peinado

John E Paff

Paul England

Eric P Traut

Bryan M Willman

Shuvabrata Ganguly

Thekkthalackal Varugis Kurien

Stephane Plante

Adrian J Oney

Brandon Allsop

Jeremy Paul Cahill

Jonathan Vines Smith

John F Wiederhirn

Kenneth D Ray

Joseph S Beda, Iii

Nathan T Lewis

Forrest Curtis Foltz

Yuqun Chen

Evan K Anderson

Suyash Sinha

Lonny Dean McMichael

Sanjeet Singh Mehat

Robert H Earhart, Iii

Jung-Ik Lee

David T McWherter

Frank Berreth

Techniques and systems for managed boot are described. A described technique includes creating, at a server system, a boot image based on one or more user-specified parameters associated with virtual machine booting, providing a first virtual drive that includes the boot image, booting a virtual machine using the boot image on the first virtual drive, and providing a second virtual drive to the virtual machine, where the virtual machine mounts the second virtual drive as a root file system.

Managed boot in a cloud system

Various aspects are disclosed herein for bounding the behavior of a non-privileged virtual machine that interacts with a device by creating a description of the device which indicates to a privileged authority (1) which operations on the device may have system-wide effects and (2) which operations have effects local to the device. The privileged authority may then permit or deny these actions. The privileged authority may also translate these actions into other actions with benign consequences.

Configuration space virtualization

Systems and methods are provided, whereby partitions may become enlightened and discover the presence of a hypervisor. Several techniques of hypervisor discovery are discussed, such as detecting the presence of virtual processor registers (e.g. model specific registers or special-purpose registers) or the presence of virtual hardware devices. Upon discovery, information (code and/or data) may be injected in a partition by the hypervisor, whereby such injection allows the partition to call the hypervisor. Moreover, the hypervisor may present a versioning mechanism that allows the partition to match up the version of the hypervisor to its virtual devices. Next, once code and/or data is injected, calling conventions are established that allow the partition and the hypervisor to communicate, so that the hypervisor may perform some operations on behalf of the partition. Four exemplary calling conventions are considered: restartable instructions, a looping mechanism, shared memory transport, and synchronous or asynchronous processed packets. Last, cancellation mechanisms are considered, whereby partition requests may be cancelled.

Systems and methods for hypervisor discovery and utilization

This document describes tools capable of making a portion of operating-system memory associated with a protection agent unalterable or inaccessible from an operating-system privilege mode. In some embodiments, these tools are capable of creating a protection-agent privilege mode by requesting that a virtual machine monitor protect this portion of operating-system memory. In other embodiments, these tools are capable of creating the protection-agent privilege mode by virtualizing a physical processor into multiple virtual processors, at least one of which is a protection-agent virtual processor designed to run the protection agent. By making this portion of operating-system memory unalterable or inaccessible from the operating-system privilege mode, the protection agent may be less vulnerable to attacks by entities operating within the operating-system privilege mode.

Protection agents and privilege modes

Various operations are disclosed for improving the operational efficiency of register handling in a virtualized environment. Some infrequently accessed software managed registers are managed lazily when switching contexts between virtual processors. The states of those registers are not saved on exit from a guest or restored on entry to the guest. Rather, guest accesses to those registers are intercepted. For some frequently accessed registers, register states are saved or restored only upon exit from a hypervisor to a different guest that that from which the hypervisor was entered. For enable-flag-gated registers, updates to a physical register value are not made unless the register is enabled. A shadow register cache may be used to speed accesses to some registers. When a shadowed register is modified, the new value is cached as a shadow copy in RAM and subsequent reads of the register are taken from the shadow copy.

Efficient context switching of virtual processors by managing physical register states in a virtualized environment

The launching of a hypervisor after there is already a running operating system. The operating system itself may launch the hypervisor. The running operating system may be used instead of the hypervisor to discover the physical resources running on the computing system. Other operating systems or operating system instances may be launched after the hypervisor is operational.

Launching hypervisor under running operating system

Plural guest operating systems run on a computer, where a security kernel enforces a policy of isolation among the guest operating systems. An exclusion vector defines a set of pages that cannot be accessed by direct memory access (DMA) devices. The security kernel enforces an isolation policy by causing certain pages to be excluded from direct access. Thus, device drivers in guest operating systems are permitted to control DMA devices directly without virtualization of those devices, while each guest is prevented from using DMA devices to access pages that the guest is not permitted to access under the policy.

Enforcing isolation among plural operating systems

Mechanisms are disclosed herein that manage operations in virtual machine environments. A first partition can have a proxy driver object corresponding to a driver object in a second partition. The driver object can control a physical device, but because of the proxy driver object, the first partition can retain some measure of control over the physical device. The driver object can be surrounded by a first filter object beneath it, and a second filter object above it. The first filter object can provide interfaces to the driver object so that the driver object can perform various bus-related functionalities; and, the second filter object can receive redirected instructions from the first partition and provide them to the driver object, and intercept any instructions originating from within the second partition, such that if these instructions are inconsistent with policies set in the first partition, they can be manipulated.

Distributed hardware state management in virtual machines

Communication of information via an in-band channel using a trusted configuration space is provided. The introduction of using a trusted configuration space associated with a computer bus system, such as PCI Express&#xae; (PCIe&#x2122;), for example, enables the design of trusted computing platforms capable of providing compliant devices with assurance that their trusted configuration registers can only be accessed by software running in the trusted software environment. Establishing device trust in the software that initiates trusted configuration requests makes it possible to secure and control access to certain secret, sensitive, or personally-identifiable information these devices may contain (e.g., a uniquely-identifying public key, as described above or certificate which the device provides for revocation purposes).

Communication of information via an in-band channel using a trusted configuration space

A machine readable description of a new feature of a processor is provided by a processor vendor. Control code executing on a processor, such as a traditional operating system kernel, a partitioning kernel, or the like can be programmed to receive the description of the feature and to use information provided by the description to detect, enable and manage operation of the new feature.

Methods for describing processor features

The present invention provides for trusted side-band communications between components in a computer system, so that use of the system bus may be avoided. Two components may be connected by means other than a bus (e.g., an infrared port, a wire, an unused pin, etc.), whereby these components may communicate without the use of the system bus. The non-bus communication channel may be referred to as 'side-band.' The side-band channel may be used to communicate information that might identify the user's hardware (e.g., a public key) or other information that the user may not want to be easily intercepted by the public at large. Communication over the side-band channel may also be used to verify that the participants in a communication are within a defined positional relationship to each other.

Communication of information via a side-band channel, and use of same to verify positional relationship

A method is provided for a processor of a computing device to obtain a trusted identification of a hardware peripheral of the computing device, for the computing device and the peripheral to derive a set of shared keys, and for the processor to send trusted data to the peripheral.

Trusted path for transmitting content thereon

A system and method to facilitate communication between an associated bus, such as employs a standard bus protocol, and a connector to which a removable SFF device can be attached. A desired operating mode is selected based on the device attached at the connector, such as either to pass the protocol between the bus and device generally unchanged or to implement suitable protocol conversion for such communication. Thus, by configuring the SFF device to appear as device currently supported by the bus, the SFF device can operate at the connector with native operating system support.

System and method to facilitate native use of small form factor devices

intercepted in order to determine if the modification will result in an aperture memory address mapping to a region of trusted memory. If it is determined that the GART modification will not result in a mapping to a portion of trusted memory, then the GART modification is allowed. If it is determined that the GART modification will result in a mapping to a portion of trusted memory, then the modification is not allowed and an alarm is raised in the system.

Growing community of inventors

Monroe, WA, United States of America

Andrew John Thornton