Advanced Micro Devices Corporation

Other

Globalfoundries Inc.

Michael T Clark

Scott A White

Kevin John McGrath

Jay E Fleischman

David Edward Kroesche

Jelena Ilic

John Kalamatianos

William Alexander Hughes

Paul J Moyer

Eric W Mahurin

Benjamin T Sander

Jagadish B Kotra

William Louie Walker

Marius Evers

Gerald David Zuraski, Jr

Kelvin Domnic Goveas

Michael A Filippo

Chetana Nagendra Keltcher

James B Keller

David S Christie

Frederick Daniel Weber

Maurice B Steinman

Derrick R Meyer

Vilas K Sridharan

Edward J McLellan

John M King

Philip Enrique Madrid

David Francis Tobias

David N Suggs

William Kurt Lewchuk

Ashraf Ahmed

Arun Radhakrishnan

Matthew Andrew Rafacz

Dean A Liberty

Matthew M Crum

Michael Thomas Dibrino

Thaddeus Fortenberry

Timothy J Wood

Michael E Tuuk

Hongwen Gao

Wallace Paul Montgomery

Bruce R Holloway

Greg Smaus

Mahesh Subramony

Steven Beigelmacher

Syed Faisal Ahmed

Cristian Constantinescu

Keith Cronin

James R Williamson

Wilbur L Clark

Kelvin Goveas

Processor-guided execution of offloaded instructions using fixed function operations is disclosed. Instructions designated for remote execution by a target device are received by a processor. Each instruction includes, as an operand, a target register in the target device. The target register may be an architected virtual register. For each of the plurality of instructions, the processor transmits an offload request in the order that the instructions are received. The offload request includes the instruction designated for remote execution. The target device may be, for example, a processing-in-memory device or an accelerator coupled to a memory.

Processor-guided execution of offloaded instructions using fixed function operations

A method of operating a processing unit includes, in response to detecting that the processing unit is operating in a voltage limited state, calculating a set of headroom values by calculating a headroom value for each operational constraint in a set of operational constraints of the processing unit, based on the calculated set of headroom values, selecting from a set of performance features a subset of one or more performance features for enabling in the processing unit, and enabling the selected subset of performance features in the processing unit.

Enabling performance features for voltage limited processors

Methods, devices, and systems for determining an address in a physical memory which corresponds to a virtual address using a skewed-associative translation lookaside buffer (TLB) are described. A virtual address and a configuration indication are received using receiver circuitry. A physical address corresponding to the virtual address is output if a TLB hit occurs. A first subset of a plurality of ways of the TLB is configured to hold a first page size. The first subset includes a number of the ways based on the configuration indication. A physical address corresponding to the virtual address is retrieved from a page table if a TLB miss occurs, and at least a portion of the physical address is installed in a least recently used way of a subset of a plurality of ways the TLB, determined according to a replacement policy based on the configuration indication.

Configurable skewed associativity in a translation lookaside buffer

Systems, apparatuses, and methods for implementing a hardware enforcement mechanism to enable platform-specific firmware visibility into an error state ahead of the operating system are disclosed. A system includes at least one or more processor cores, control logic, a plurality of registers, platform-specific firmware, and an operating system (OS). The control logic allows the platform-specific firmware to decide if and when the error state is visible to the OS. In some cases, the platform-specific firmware blocks the OS from accessing the error state. In other cases, the platform-specific firmware allows the OS to access the error state such as when the OS needs to unmap a page. The control logic enables the platform-specific firmware, rather than the OS, to make decisions about the replacement of faulty components in the system.

Platform first error handling

Systems, apparatuses, and methods for tracking system resource utilization of guest virtual machines (VMs). Counters may be maintained to track resource utilization of different system resources by different guest VMs executing on the system. When a guest VM initiates execution, stored values may be loaded into the resource utilization counters. While the guest VM executes, the counters may track the resource utilization of the guest VM. When the guest VM terminates execution, the counter values may be written to a virtual machine control block (VMCB) corresponding to the guest VM. Scaling factors may be applied to the counter values to normalize the values prior to writing the values to the VMCB. A cloud computing environment may utilize the tracking mechanisms to guarantee resource utilization levels in accordance with users' service level agreements.

Mechanism for resource utilization metering in a computer system

A method of implementing a mask load or mask store instruction by a processor is provided. The method may include receiving the mask load or mask store instruction, a location of a memory operand and a location of corresponding mask bits associated with the memory operand, breaking the received memory operand into a plurality of sub-operands and executing the mask load or mask store instruction on each of the plurality of sub-operands using a fastpath operation or using microcode, wherein the respective mask load or mask store instruction loads or stores each of the plurality of sub-operands based upon the corresponding mask bits.

Processor and method implemented by a processor to implement mask load and store instructions

A data processing device employs a first translation look-aside buffer (TLB) to translate virtual addresses to physical addresses. If a virtual address to be translated is not located in the first TLB, the physical address is requested from a set of page tables. When the data processing device is in a hypervisor mode, a second TLB is accessed in response to the request to access the page tables. If the virtual address is located in the second TLB, the hypervisor page tables are bypassed and the second TLB provides a physical address or information to access another table in the set of page tables. By bypassing the hypervisor page tables, the time to translate an address in the hypervisor mode is reduced, thereby improving the efficiency of the data processing device.

Address translation with multiple translation look aside buffers

In an embodiment, a microcode unit for a processor is contemplated. The microcode unit comprises a microcode memory storing a plurality of microcode routines executable by the processor, wherein each microcode routine comprises two or more microcode operations. Coupled to the microcode memory, the sequence control unit is configured to control reading microcode operations from the microcode memory to be issued for execution by the processor. The sequence control unit is configured to stall issuance of microcode operations forming a body of a loop in a first routine of the plurality of microcode routines until a loop counter value that indicates a number of iterations of the loop is received by the sequence control unit.

Precise counter hardware for microcode loops

A processor receives a command via a sideband interface on the processor to read processor state information, e.g., CPUID information. The sideband interface provides the command information to a microcode engine in the processor that executes the command to retrieve the designated processor state information at an appropriate instruction boundary and retrieves the processor state information. That processor information is stored in local buffers in the sideband interface to avoid modifying processor state. After the microcode engine completes retrieval of the information and the sideband interface command is complete, execution returns to the normal flow in the processor. Thus, the processor state information may be obtained non-destructively during processor runtime.

Non-destructive sideband reading of processor state information

A system and method for copying and initializing a block of memory. To copy several data entities from a source region of memory to a destination region of memory, an instruction may copy each data entity one at a time. If an aggregate condition is determined to be satisfied, multiple data entities may be copied simultaneously. The aggregate condition may rely on an aggregate data size, the size of the data entities to be copied, and the alignment of the source and destination addresses.

Data movement and initialization aggregation

A single test access port, such as a JTAG-based debug port may be utilized to perform debug operations on logic cores of a multi-core integrated circuit, such as a multi-core processor. The shared debug port may respond to a particular command to enter a debugging mode and may be configured to forward all commands and data to a debugging controller of the integrated circuit during debugging. A mask register may be used to indicate which logic cores of the multi-core integrated circuit should be debugged. Additionally, custom debugging commands may include mask or core select fields to indicate which logic cores should be affected by the particular command. Debugging mode may be initialized for one or more logic cores either externally, such as be asserted a DBREQ signal, or internally, such as by configuring one or more breakpoints.

Multi-core integrated circuit with shared debug port

A technique of operating a processor includes determining whether a floating point unit (FPU) of the processor is to operate in a full-bit mode or a reduced-bit mode. An instruction is fetched and the instruction is decoded into a single operation, when the full-bit mode is indicated, or multiple operations, when the reduced-bit mode is indicated.

Processor with power saving reconfigurable floating point unit decoding an instruction to single full bit operation or multiple reduced bit operations

A processing node that is integrated onto a single integrated circuit chip includes a first processor core and a second processor core. The processing node also includes an operating system executing on either of the first processor core and the second processor core. The operating system may monitor a current utilization of the first processor core and the second processor core. The operating system may cause the first processor core to operate at performance level that is lower than a system maximum performance level and the second processor core to operate at performance level that is higher than the system maximum performance level in response to detecting the first processor core operating below a utilization threshold.

Increasing workload performance of one or more cores on multiple core processors

A method of controlling write operations to a non-renamed register space includes receiving a write operation to a given register within the non-renamed register space. The method also includes determining whether a pending write operation to the given register exists. In response to determining that the pending write operation to the given register exists, the method includes blocking the write operation to the given register from being scheduled. However, in response to determining that the pending write operation to the given register does not exist, the method includes allowing the write operation to the given register to be scheduled. Further, if the pending write operation to the given register does not exist, the method includes allowing a subsequent write operation to a different register within the non-renamed register space to be scheduled.

Controlling writes to non-renamed register space in an out-of-order execution microprocessor

Growing community of inventors

Austin, TX, United States of America

Michael T Clark