Other

Transmeta Corporation

Intel Corporation

H Peter Anvin

Guillermo Juan Rozas

John P Banning

Gilbert A Neiger

Alexander Klaiber

Linus Torvalds

Asit K Mallick

Marc Fleischmann

David A Dunn

Jason Burke Wilson

Andrew M Shaw

Jason W Brandt

Barry E Huntley

Scott Dion Rodgers

David Pardo Keppel

Zongjian Chen

Godfrey P D'Souza

Adriaan Van De Ven

Robert Bedichek

Andrew Read

Keith Klayman

Alex Klaiber

Sameer Halepete

Thomas R Lawrence

James B Crossland

Parag Gupta

Andrew Morgan

Vedvyas Shanbhogue

Rajesh M Sankaran

Raghunandan Makaram

Michael Mishaeli

Deepak K Gupta

Benjamin C Chaffin

Richard C Johnson

Vincent J Zimmer

Michael A Rothman

Martin Guy Dixon

Naveen Kumar

Baiju V Patel

Naveen Neelakantam

Xiaoning Li

Paul S Serris

James S Coke

Toby Opferman

Gopinatth Selvaraje

Sebastian Winkel

Kameswar Subramaniam

David C Estrada

Atul A Khare

Nicholas J Yoke

Brian L Vajda

Ori Lempel

Michael Neilly

Christopher M Russell

Nirajan L Cooray

Robert Bedicheck

Benjamin Gribstad

In one embodiment, a processor includes: a front end circuit to fetch and decode a read list instruction, the read list instruction to cause storage to a memory of a software-provided list of processor state information; and an execution circuit coupled to the front end circuit. The execution circuit, in response to the decoded read list instruction, is to read the processor state information stored in the processor and store each datum of the processor state information into an entry of a data table in the memory. Other embodiments are described and claimed.

System, apparatus and methods for performant read and write of processor state information responsive to list instructions

A processor of an aspect includes a decode unit to decode a user-level suspend thread instruction that is to indicate a first alternate state. The processor also includes an execution unit coupled with the decode unit. The execution unit is to perform the instruction at a user privilege level. The execution unit in response to the instruction, is to: (a) suspend execution of a user-level thread, from which the instruction is to have been received; (b) transition a logical processor, on which the user-level thread was to have been running, to the indicated first alternate state; and (c) resume the execution of the user-level thread, when the logical processor is in the indicated first alternate state, with a latency that is to be less than half a latency that execution of a thread can be resumed when the logical processor is in a halt processor power state.

Methods, apparatus, and instructions for user-level thread suspension

Embodiments of an invention for protecting supervisor mode information are disclosed. In one embodiment, an apparatus includes a storage location, instruction hardware, execution hardware, and control logic. The storage location is to store an indicator to enable supervisor mode information protection. The instruction hardware is to receive an instruction to access supervisor mode information. The execution hardware is to execute the instruction. The control logic is to prevent execution of the instruction if supervisor mode information protection is enabled and a current privilege level is less privileged than a supervisor mode.

Protecting supervisor mode information

An apparatus and method for efficiently managing shadow stacks. For example, one embodiment of a processor comprises: a plurality of registers to store a plurality of shadow stack pointers (SSPs); event processing circuitry to select a first SSP of the plurality of SSPs from a first register of the plurality of registers responsive to receipt of a first event associated with a first event priority level, the first SSP usable to identify a top of a first shadow stack; verification and utilization checking circuitry to determine whether the first SSP has been previously verified, wherein if the first SSP has not been previously verified then initiating a set of atomic operations to verify the first SSP and confirm that the first SSP is not in use, the set of atomic operations using a locking operation to lock data until the set of atomic operations are complete.

Shadow stack ISA extensions to support fast return and event delivery (FRED) architecture

An apparatus and method for efficiently managing shadow stacks. For example, one embodiment of a processor comprises: a plurality of registers to store a plurality of shadow stack pointers (SSPs), each SSP associated with a different event priority; event processing circuitry to select a first SSP of the plurality of SSPs from a first register of the plurality of registers responsive to receipt of a first event associated with a first event priority level, the first SSP usable to identify a top of a first shadow stack; verification and utilization checking circuitry to determine whether the first SSP has been previously verified, wherein if the first SSP has not been previously verified then initiating a set of atomic operations to verify the first SSP and confirm that the first SSP is not in use, the set of atomic operations using a locking operation to lock data until the set of atomic operations are complete, and wherein if the first SSP has been previously verified, then re-verifying the first SSP and confirming that the first SSP is not in use without using the locking operation.

Methods, apparatus, and instructions for user level thread suspension

A processor of an aspect includes a decode unit to decode an exception handler return instruction. The processor also includes an exception handler return execution unit coupled with the decode unit. The exception handler return execution unit, responsive to the exception handler return instruction, is to not configure the processor to enable delivery of a subsequently received nonmaskable interrupt (NMI) to an NMI handler if an exception, which corresponds to the exception handler return instruction, was taken within the NMI handler. The exception handler return execution unit, responsive to the exception handler return instruction, is to configure the processor to enable the delivery of the subsequently received NMI to the NMI handler if the exception was not taken within the NMI handler. Other processors, methods, systems, and instructions are disclosed.

Avoiding premature enabling of nonmaskable interrupts when returning from exceptions

The present disclosure is directed to flexible bootstrap code architecture. A device may comprise equipment for operating the device and an operating system (OS) for operating the equipment. A boot module may also be included in the device to execute boot operations. At least one flexible boot (FB) module in the boot module may interact with the equipment and/or OS during the boot operations to cause the boot operations to become device-specific. An example boot module may comprise a plurality of FB modules. An example FB module may verify a device/chipset identification and may control the boot operations based on the identification. Other example FB modules may select resources to load based on an OS type, may provide a boot configuration table location for use in OS runtime boot configuration or may load variables from a preload variable directory for use in configuring boot operations.

Flexible bootstrap code architecture

Instructions and logic provide memory key protection functionality. Embodiments include a processor having a register to store a memory protection field. A decoder decodes an instruction having an addressing form field for a memory operand to specify one or more memory addresses, and a memory protection key. One or more execution units, responsive to the memory protection field having a first value and to the addressing form field of the decoded instruction having a second value, enforce memory protection according to said first value of the memory protection field, using the specified memory protection key, for accessing the one or more memory addresses, and fault if a portion of the memory protection key specified by the decoded instruction does not match a stored key value associated with the one or more memory addresses.

Instructions and logic to provide memory access key protection functionality

An example processing system may comprise: a lower stack bound register configured to store a first memory address, the first memory address identifying a lower bound of a memory addressable via a stack segment; an upper stack bound register configured to store a second memory address, the second memory address identifying an upper bound of the memory addressable via the stack segment; and a stack bounds checking logic configured to detect unauthorized stack pivoting, by comparing a memory address being accessed via the stack segment with at least one of the first memory address and the second memory address.

Systems and methods for preventing unauthorized stack pivoting

A processor includes a processor core to execute a first translated instruction translated from a first instruction stored in first page of a memory. The processor also includes a translation indicator agent (XTBA) to store a first translation indicator that is read from a physical map (PhysMap) in the memory. In an embodiment, the first translation indicator is to indicate whether the first page has been modified after the first instruction is translated. Other embodiments are described as claimed.

Handling of binary translated self modifying code and cross modifying code

A method for caching attribute data for matching attributes with physical addresses. The method includes storing a plurality of attribute entries in a memory, wherein the memory is configured to provide at least one attribute entry when accessed with a physical address, and wherein the attribute entry provided describes characteristics of the physical address.

Method and system for caching attribute data for matching attributes with physical addresses

Growing community of inventors

San Jose, CA, United States of America

H Peter Anvin