Nvidia Corporation

Poornachandra Rao

Steven James Heinrich

Shirish C Gadre

Jack H Choquette

Xiaogang Qiu

Omkar Paranjape

Ramesh Jandhyala

John H Edmondson

Olivier Giroux

Ronny Krashinsky

Manan Patel

Andrew Robert Kerr

Mark Alan Gebhart

Alan Kaatz

Michael Siu

Jeffrey A Bolz

Eric T Anderson

Jonathan J Dunaisky

Jerome Francis Duluk, Jr

Henry Packard Moreton

James Leroy Deming

Rui Manuel Bastos

Yury Y Uralsky

Patrick Richard Brown

Alexander Lev Minkin

Bryon S Nordquist

Joel James McCormack

Andrew J Tao

Christian Amsinck

Amanpreet Grewal

A technique for block data transfer is disclosed that reduces data transfer and memory access overheads and significantly reduces multiprocessor activity and energy consumption. Threads executing on a multiprocessor needing data stored in global memory can request and store the needed data in on-chip shared memory, which can be accessed by the threads multiple times. The data can be loaded from global memory and stored in shared memory using an instruction which directs the data into the shared memory without storing the data in registers and/or cache memory of the multiprocessor during the data transfer.

Techniques for efficiently transferring data to a processor

A unified cache subsystem includes a data memory configured as both a shared memory and a local cache memory. The unified cache subsystem processes different types of memory transactions using different data pathways. To process memory transactions that target shared memory, the unified cache subsystem includes a direct pathway to the data memory. To process memory transactions that do not target shared memory, the unified cache subsystem includes a tag processing pipeline configured to identify cache hits and cache misses. When the tag processing pipeline identifies a cache hit for a given memory transaction, the transaction is rerouted to the direct pathway to data memory. When the tag processing pipeline identifies a cache miss for a given memory transaction, the transaction is pushed into a first-in first-out (FIFO) until miss data is returned from external memory. The tag processing pipeline is also configured to process texture-oriented memory transactions.

Unified cache for diverse memory traffic

A texture processing pipeline is configured to store decoded texture data within a cache unit in order to expedite the processing of texture requests. When a texture request is processed, the texture processing pipeline queries the cache unit to determine whether the requested data is resident in the cache. If the data is not resident in the cache unit, a cache miss occurs. The texture processing pipeline then reads encoded texture data from global memory, decodes that data, and writes different portions of the decoded memory into the cache unit at specific locations according to a caching map. If the data is, in fact, resident in the cache unit, a cache hit occurs, and the texture processing pipeline then reads decoded portions of the requested texture data from the cache unit and combines those portions according to the caching map.

Approach to caching decoded texture data with variable dimensions

One embodiment of the present invention includes a memory management unit (MMU) that is configured to manage sparse mappings. The MMU processes requests to translate virtual addresses to physical addresses based on page table entries (PTEs) that indicate a sparse status. If the MMU determines that the PTE does not include a mapping from a virtual address to a physical address, then the MMU responds to the request based on the sparse status. If the sparse status is active, then the MMU determines the physical address based on whether the type of the request is a write operation and, subsequently, generates an acknowledgement of the request. By contrast, if the sparse status is not active, then the MMU generates a page fault. Advantageously, the disclosed embodiments enable the computer system to manage sparse mappings without incurring the performance degradation associated with both page faults and conventional software-based sparse mapping management.

Managing memory regions to support sparse mappings

Approaches are disclosed for performing memory access operations in a texture processing pipeline having a first portion configured to process texture memory access operations and a second portion configured to process non-texture memory access operations. A texture unit receives a memory access request. The texture unit determines whether the memory access request includes a texture memory access operation. If the memory access request includes a texture memory access operation, then the texture unit processes the memory access request via at least the first portion of the texture processing pipeline, otherwise, the texture unit processes the memory access request via at least the second portion of the texture processing pipeline. One advantage of the disclosed approach is that the same processing and cache memory may be used for both texture operations and load/store operations to various other address spaces, leading to reduced surface area and power consumption.

Load/store operations in texture hardware

Growing community of inventors

Cedar Park, TX, United States of America

Poornachandra Rao