derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
65 lines
2.6 KiB
ReStructuredText
65 lines
2.6 KiB
ReStructuredText
.. meta::
|
|
:description: rocPRIM documentation and API reference library
|
|
:keywords: rocPRIM, ROCm, API, documentation
|
|
|
|
.. _glossary:
|
|
|
|
********************************************************************
|
|
Glossary of rocPRIM terms
|
|
********************************************************************
|
|
|
|
This glossary is to help users understand the basic concepts or terminologies used in the rocPRIM library.
|
|
|
|
.. glossary::
|
|
:sorted:
|
|
|
|
Wavefront
|
|
A group of threads that runs using the single instruction, multiple thread (SIMT) model.
|
|
|
|
Wave
|
|
See :term:`wavefront<Wavefront>`. rocPRIM uses "warp", "wave", and "wavefront" interchangeably.
|
|
|
|
Work-item
|
|
A work-item is the smallest unit of parallel execution. A work-item runs a single independent instruction stream on a single data element.
|
|
|
|
Thread
|
|
See :term:`work-item<Work-item>`. rocPrim uses "thread" and "work-item" interchangeably.
|
|
|
|
Warp
|
|
Alternate term for a :term:`wavefront<Wavefront>`. rocPRIM uses "warp", "wave", and "wavefront" interchangeably.
|
|
|
|
Hardware warp size
|
|
The number of threads in a warp as defined by the hardware. On AMD GPUs, the warp size can be either thirty-two (32) or sixty-four (64) threads.
|
|
|
|
Logical warp size
|
|
The number of threads in a warp as defined by the user. This can be equal to or less than the size of the hardware warp size.
|
|
|
|
Block
|
|
See :term:`tile<Tile>`. rocPRIM uses "block" and "tile" interchangeably.
|
|
|
|
Stride
|
|
The number of threads per block.
|
|
|
|
Tile
|
|
A group of warps that run on the same streaming multiprocessor (SM). Threads in the block can be indexed using one dimension, {X}, two dimensions, {X, Y}, or three dimensions, {X, Y, Z}. In rocPRIM the tile size is always the same as the block size.
|
|
|
|
Grid
|
|
A group of blocks that all run the same kernel call.
|
|
|
|
Warp ID
|
|
The identifier of the warp within a block. A warp's warp ID is guaranteed to be unique.
|
|
|
|
Thread ID
|
|
The identifier of the thread within a block.
|
|
|
|
Lane ID
|
|
The identifier of the thread within the warp.
|
|
|
|
Flat ID
|
|
The flattened block or thread idex. The flat ID is a one-dimensional index created from two-dimensional or three-dimensional indices. For example the flat ID of a two-dimensional thread ID {X, Y} in a two-dimensional ``128x4`` block is ``Y*128*X``.
|
|
|
|
SIMT
|
|
See :term:`Single-Instruction, Multi-Thread`.
|
|
|
|
Single-Instruction, Multi-Thread
|
|
Single-instruction, multi-thread (SIMT) is a parallel computing model where all the :term:`work-items<work-item>` within a :term:`wavefront` run the same instruction on different data. |