Application programming interface to share memory between groups of blocks of threads

Patent metadata

Field	Value
Publication number	US-12578993-B2
Application number	US-202217955175-A
Country	US
Kind code	B2
Filing date	Sep 28, 2022
Priority date	Jul 29, 2022
Publication date	Mar 17, 2026
Grant date	Mar 17, 2026

Abstract

Apparatuses, systems, and techniques to execute CUDA programs. In at least one embodiment, an application programming interface is performed to cause memory to be shared between two or more groups of blocks of threads.

First claim

What is claimed is:

1 . A processor comprising:
one or more circuits to perform an application programming interface (API) to cause memory to be shared between two or more groups of blocks of threads.

2 . The processor of claim 1 , wherein the two or more groups of blocks of threads are groups within a software kernel.

3 . The processor of claim 1 , wherein the two or more groups of blocks of threads are groups within a grid of blocks of threads.

4 . The processor of claim 1 , wherein the two or more groups of blocks of threads are partitions of a partitioning of blocks of threads.

5 . The processor of claim 1 , wherein the memory is memory of a graphics processing unit (GPU).

6 . The processor of claim 1 , wherein the API is to cause at least one thread of one group of blocks of threads to be able to access a memory location accessible to at least one thread of another group of blocks of threads.

7 . The processor of claim 1 , wherein the API is to cause at least one thread of one group of blocks of threads to provide a memory address of the memory to be shared to the two or more groups of blocks of threads.

8 . The processor of claim 1 , wherein the API is to cause at least one thread of a first group of blocks of threads to calculate a memory address of the memory to be shared based, at least in part, on an identifier of a block of the first group of blocks of threads.

9 . The processor of claim 1 , wherein the API is to cause at least one thread of a first group of blocks of threads to calculate a memory address of the memory to be shared based, at least in part, on an identifier of the first group of blocks of threads.

10 . A computer-implemented method comprising:
performing an application programming interface (API) to cause memory to be shared between two or more groups of blocks of threads.

11 . The computer-implemented method of claim 10 , wherein the two or more groups of blocks of threads are groups within a software kernel.

12 . The computer-implemented method of claim 10 , wherein the two or more groups of blocks of threads are groups within a grid of blocks of threads.

13 . The computer-implemented method of claim 10 , wherein the two or more groups of blocks of threads are partitions of a partitioning of blocks of threads.

14 . The computer-implemented method of claim 10 , wherein the memory is memory of a graphics processing unit (GPU).

15 . The computer-implemented method of claim 10 , wherein the API is to cause at least one thread of one group of blocks of threads to be able to access a memory location accessible to at least one thread of another group of blocks of threads.

16 . The computer-implemented method of claim 10 , wherein the API is to cause at least one thread of one group of blocks of threads to provide a memory address of the memory to be shared to the two or more groups of blocks of threads.

17 . The computer-implemented method of claim 10 , wherein the API is to cause at least one thread of a first group of blocks of threads to calculate a memory address of the memory to be shared based, at least in part, on an identifier of a block of the first group of blocks of threads.

18 . The computer-implemented method of claim 10 , wherein the API is to cause at least one thread of a first group of blocks of threads to calculate a memory address of the memory to be shared based, at least in part, on an identifier of the first group of blocks of threads.

19 . A computer system comprising:
one or more processors and memory storing executable instructions that, if performed by the one or more processors, are to perform an application programming interface (API) to cause memory to be shared between two or more groups of blocks of threads.

20 . The computer system of claim 19 , wherein the two or more groups of blocks of threads are groups within a software kernel.

21 . The computer system of claim 19 , wherein the two or more groups of blocks of threads are groups within a grid of blocks of threads.

22 . The computer system of claim 19 , wherein the two or more groups of blocks of threads are partitions of a partitioning of blocks of threads.

23 . The computer system of claim 19 , wherein the memory is memory of a graphics processing unit (GPU).

24 . The computer system of claim 19 , wherein the API is to cause at least one thread of one group of blocks of threads to be able to access a memory location accessible to at least one thread of another group of blocks of threads.

25 . The computer system of claim 19 , wherein the API is to cause at least one thread of one group of blocks of threads to provide a memory address of the memory to be shared to the two or more groups of blocks of threads.

26 . The computer system of claim 19 , wherein the API is to cause at least one thread of a first group of blocks of threads to calculate a memory address of the memory to be shared based, at least in part, on an identifier of a block of the first group of blocks of threads.

27 . The computer system of claim 19 , wherein the API is to cause at least one thread of a first group of blocks of threads to calculate a memory address of the memory to be shared based, at least in part, on an identifier of the first group of blocks of threads.

28 . A machine-readable medium having stored thereon a set of instructions, which if performed by one or more processors, are to perform an application programming interface (API) to cause memory to be shared between two or more groups of blocks of threads.

29 . The machine-readable medium of claim 28 , wherein the two or more groups of blocks of threads are groups within a software kernel.

30 . The machine-readable medium of claim 28 , wherein the two or more groups of blocks of threads are groups within a grid of blocks of threads.

31 . The machine-readable medium of claim 28 , wherein the two or more groups of blocks of threads are partitions of a partitioning of blocks of threads.

32 . The machine-readable medium of claim 28 , wherein the memory is memory of a graphics processing unit (GPU).

33 . The machine-readable medium of claim 28 , wherein the API is to cause at least one thread of one group of blocks of threads to be able to access a memory location accessible to at least one thread of another group of blocks of threads.

34 . The machine-readable medium of claim 28 , wherein the API is to cause at least one thread of one group of blocks of threads to provide a memory address of the memory to be shared to the two or more groups of blocks of threads.

35 . The machine-readable medium of claim 28 , wherein the API is to cause at least one thread of a first group of blocks of threads to calculate a memory address of the memory to be shared based, at least in part, on an identifier of a block of the first group of blocks of threads.

36 . The machine-readable medium of claim 28 , wherein the API is to cause at least one thread of a first group of blocks of threads to calculate a memory address of the memory to be shared based, at least in part, on an identifier of the first group of blocks of threads.

Assignees

• Nvidia Corp

Inventors

• Long Ze
• Perelygin Kyrylo
• Edwards Harold Carter
• Hirisave Chandra Shekhara Gokul Ramaswamy
• Marathe Jaydeep
• Krashinsky Ronny Meir
• Bharambe Girish Bhaskarrao

Classifications

• G06F9/505

  considering the load · CPC title

• G06F9/522

  Barrier synchronisation · CPC title

• G06F9/30072

  to perform conditional operations, e.g. using predicates or guards · CPC title

• G06F9/544

  Buffers; Shared memory; Pipes · CPC title

• G06F8/456 Primary

  Parallelism detection · CPC title