# US 8,752,064 B2 · Optimizing communication of system call requests > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,752,064 B2 - Document ID: US8752064B2 - Status: Granted - Kind: B2 - Application number: 13/307,505 - Filed: 2011-11-30 - Published: 2012-07-12 - Granted: 2014-06-10 - Earliest claimed: 2010-12-14 - Assignees: Advanced Micro Devices, Inc. - Applicants: Benjamin Thomas Sander; Michael Houston; Newton Cheung; Keith Lowery - Inventors: Benjamin Thomas Sander; Michael Houston; Newton Cheung; Keith Lowery - Web page: https://patentgrep.com/patent/US8752064B2 ## CPC - G06F 9/544: {Buffers; Shared memory; Pipes} - G06F 9/522: {Barrier synchronisation} - G06F 9/546: {Message passing systems or structures, e.g. queues} - G06T 1/20: Processor architectures; Processor configuration, e.g. pipelining - G06F 2209/509: Offload - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract Provided herein is a method for optimizing communication for system calls. The method includes storing a system call for each work item in a wavefront and transmitting said stored system calls to a processor for execution. The method also includes receiving a result to each work item in the wavefront responsive to said transmitting. ## Claims ### Claim 1 A method comprising: storing, in a corresponding single instructor multiple data (SIMD) element, information to process a system call request for each work item in a wavefront; grouping the SIMD elements as an SIMD vector to create a system call request data structure; transmitting the SIMD vector to a processor for execution; and receiving a result to each work item. ### Claim 2 (depends on 1) The method of claim 1, further comprising enqueueing the SIMD vector within a high-priority queue visible to a central processing unit (CPU). ### Claim 3 (depends on 1) The method of claim 1, wherein each SIMD element comprises a function selector, an argument list, and a memory space for the result. ### Claim 4 (depends on 1) The method of claim 1, wherein the SIMD vector includes system call requests from multiple wavefronts. ### Claim 5 A method, executed in a computer system having at least one processor, comprising: receiving a single instructor multiple data CSIMD) vector, wherein the SIMD vector comprises SIMD elements corresponding to a system call request from each work item in a wavefront; executing each system call request in each SIMD element; and transmitting, via the SIMD vector, the result of each system call to each work item in the wavefront. ### Claim 6 (depends on 5) The method of claim 5, wherein the receiving includes receiving the SIMD vector in a high priority queue visible to a graphics processing device. ### Claim 7 A system comprising: a memory configured to store, in a corresponding single instructor multiple data (SIMD), element, information to process a system call request for each work item in a wavefront, wherein the SIMD elements are grouped as an SIMD vector; and a CPU configured to: execute each system call request stored in the SIMD elements, and transmit a result of each system call request to each work item in the wavefront. ### Claim 8 (depends on 7) The system of claim 7, wherein the memory is a high-priority queue configured to enqueue the SIMD vector, the high-priority queue being visible to a CPU. ### Claim 9 (depends on 7) The system of claim 7, wherein each SIMD element comprises a function selector, an argument list, and a memory space for the result. ### Claim 10 (depends on 8) The system of claim 8, wherein the SIMD vector includes system call requests from multiple wavefronts. ### Claim 11 A system, comprising: a memory; and a CPU configured to: receive a single instruction multiple data (SIMD) vector, wherein the SIMD vector includes SIMD elements containing information to process a corresponding system call request for each work item in a wavefront, execute each system call stored in the SIMD vector, and transmit the result of each system call to each work item in the wavefront. ### Claim 12 (depends on 11) The system of claim 11, wherein the CPU is further configured to: receive the SIMD vector from a high priority queue. ### Claim 13 An article of manufacture including a computer storage device having instructions stored thereon that, when executed by a computing device, cause said computing device to perform operations comprising: storing, in a corresponding single instructor multiple data (SIMD) element, information to process a system call request for each work item in a wavefront; grouping the SIMD elements as an SIMD vector to create a system call request data structure; transmitting the SIMD vector to a processor for execution; and receiving a result to each work item in the wavefront. ### Claim 14 (depends on 13) The article of manufacture of claim 13, further comprising enqueueing the SIMD vector within a high-priority queue visible to a CPU. ### Claim 15 An article of manufacture including a computer storage device having instructions stored thereon that, when executed by a computing device, cause said computing device to perform operations comprising: receiving a single instructor multiple data (SIMD) vector, wherein the SIMD vector comprises SIMD elements corresponding to a system call request from each work item in a wavefront; executing each system call request from the SIMD vector; and transmitting the result of each system call to each work item in the wavefront. ### Claim 16 (depends on 15) The article of manufacture of claim 15, wherein the receiving includes receiving the SIMD vector in a high priority queue visible to a graphics processing device. ### Claim 17 A computer storage device having instructions recorded thereon that, when executed by a computing device, cause the computing device to perforin operations comprising: storing, in a corresponding single instructor multiple data (SIMD) element, information to process a system call request for each work item in a wavefront; transmitting said stored system calls, via an SIMD vector, to a processor for execution, wherein the SIMD vector comprises the SIMD elements grouped as a single data structure; and responsive to said transmitting, receiving a result to each work item in the wavefront. ### Claim 18 (depends on 17) The computer storage device of claim 17, further comprising enqueueing the SIMD vector within a high-priority queue visible to a central processing unit (CPU). ### Claim 19 A computer storage device having instructions recorded thereon that, when executed by a computing device, cause the computing device to perform operations comprising: receiving a single instructor multiple data (SIMD) vector, wherein the SIMD vector comprises SIMD elements corresponding to a system call request from each work item in a wavefront; executing each system call request from the SIMD vector; and transmitting the result of each system call to each work item in the wavefront. ### Claim 20 (depends on 19) The computer storpe device of claim 19, wherein the SIMD vector is received from a high priority queue visible to the graphics processing device.