# US 2018/0321936 A1 · Method and system to provide user-level multithreading

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2018/0321936 A1
- Document ID: US20180321936A1
- Status: Published application
- Kind: A1
- Application number: 15/943,609
- Filed: 2018-04-02
- Published: 2018-11-08
- Earliest claimed: 2004-03-31
- Assignees: Intel Corporation
- Applicants: Intel Corporation
- Inventors: Edward Grochowski; Hong Wang; John P. Shen; Perry H. Wang; Jamison D. Collins; James Held; Partha Kundu; Raya Leviathan; Tin-Fook Ngai
- Web page: https://patentgrep.com/patent/US20180321936A1

## CPC

- G06F 9/30003: {Arrangements for executing specific machine instructions}
- G06F 9/30087: {Synchronisation or serialisation instructions}
- G06F 9/3009: {Thread control instructions}
- G06F 9/30101: {Special purpose registers}
- G06F 9/3013: {according to data content, e.g. floating-point registers, address registers}
- G06F 9/384: {Register renaming}
- G06F 9/3851: {from multiple instruction streams, e.g. multistreaming}
- G06F 9/3888: {controlled by a single instruction for multiple threads [SIMT] in parallel}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method and system to provide user-level multithreading are disclosed. The method according to the present techniques comprises receiving programming instructions to execute one or more shared resource threads (shreds) via an instruction set architecture (ISA). One or more instruction pointers are configured via the ISA; and the one or more shreds are executed simultaneously with a microprocessor, wherein the microprocessor includes multiple instruction sequencers.

## Claims

### Claim 1

A processor comprising:
a first plurality of cores having first performance and power characteristics;
a second plurality of cores having second performance and power characteristics, different from the first performance and power characteristics;
each of the first and second plurality of cores to support a 64-bit instruction set architecture (ISA) and a 32-bit ISA,
each core of the first and second plurality of cores comprising 64-bit architectural registers to be used when executing both 64-bit ISA instructions of a first thread and 32-bit ISA instructions of a second thread, each core to map 32-bit architectural registers of the 32-bit ISA to one or more of the 64-bit architectural registers during execution of the 32-bit ISA instructions of the second thread;
a first plurality of physical 128-bit registers of a first core to store packed data elements for single instruction multiple data (SIMD) operations of the first thread;
a second plurality of physical 128-bit registers of a second core to store packed data elements for single instruction multiple data (SIMD) operations of the second thread;
a first plurality of control registers of the first core to store one or more variables related to a current execution state of the first thread; and
a second plurality of control registers of the second core to store one or more variables related to a current execution state of the second thread.

### Claim 2 (depends on 1)

The processor of claim 1 wherein the first core is to switch from executing the 64-bit ISA instructions of the first thread to executing 32-bit ISA instructions of another thread responsive to execution of a first instruction.

### Claim 3 (depends on 1)

The processor of claim 1 further comprising:
execution circuitry to execute a first instruction of the first thread to cause the first thread to wait until a second instruction is executed before resuming execution of the first thread.

### Claim 4 (depends on 3)

The processor of claim 3, wherein the second instruction is from the second thread.

### Claim 5 (depends on 4)

The processor of claim 4 wherein the first thread is to resume execution following execution of the second instruction, using at least one value accessed and/or modified during execution of the second thread.

### Claim 6 (depends on 5)

The processor of claim 5 wherein the at least one value is to be stored by the second thread in a first memory location or a first register shared by the first and second threads.

### Claim 7 (depends on 6)

The processor of claim 6 wherein a shared register semaphore is implemented to share the first register and/or a shared memory semaphore is implemented to share the first memory location.

### Claim 8 (depends on 2)

The processor of claim 2 wherein the execution circuitry is further to execute a third instruction of a third thread to cause the third thread to wait until a fourth instruction of a fourth thread is executed before resuming execution of the third thread.

### Claim 9 (depends on 8)

The processor of claim 8, wherein the third thread is to resume execution following execution of the fourth instruction, using at least one value accessed during execution of the fourth thread.

### Claim 10 (depends on 9)

The processor of claim 9 wherein, in response to a context switch, the execution circuitry is to execute a context save instruction to save a state associated with the third thread in a region of memory allocated to the third thread.

### Claim 11 (depends on 10)

The processor of claim 10 wherein the execution circuitry is to execute a context restore instruction to restore the state and to continue execution of the third thread.

### Claim 12 (depends on 11)

The processor of claim 11 wherein the state comprises a first plurality of values associated with the third thread that are not shared with one or more other threads and a second plurality of values shared with the one or more other threads, the one or more other threads including the fourth thread.

### Claim 13 (depends on 12)

The processor of claim 12 wherein the second plurality of values are shared within a memory subsystem including at least one cache.

### Claim 14 (depends on 1)

The processor of claim 1 wherein the first and second plurality of control registers include a first and second control register, respectively, to store first and second base addresses of first and second regions in memory, respectively, for storing first and second interrupt data associated with the first and second threads, respectively.

### Claim 15 (depends on 1)

The processor of claim 1 wherein each of the first and second plurality of cores comprises circuitry to implement a virtual execution environment comprising a virtual machine monitor (VMM).

### Claim 16 (depends on 15)

The processor of claim 15 wherein the VMM is to detect an execution of a privileged instruction by a virtual machine and to service the privileged instruction.

### Claim 17 (depends on 1)

The processor of claim 1 wherein the first plurality of cores comprise the first core and the second plurality of cores comprise the second core.

### Claim 18 (depends on 1)

The processor of claim 1 wherein the first plurality of cores comprise the second core and the second plurality of cores comprise the first core.
