cp.reduce.async.bulk GPU Virtual ISA NVIDIA

cp.reduce.async.bulk Data Movement and Conversion Instructions

cp.reduce.async.bulk{.sem.scope}.dst.src.completion_mechanism.redOp.type

cp.reduce.async.bulk is a non-blocking instruction which initiates an asynchronous reduction operation on an array of memory locations specified by the destination address operand dstMem with the source array whose location is specified by the source address operand srcMem.

Encoding

PTX is a virtual instruction set. It has no single, stable native binary encoding - the compiler lowers this instruction to different native machine code depending on the selected NVIDIA target architecture (compute capability). This page intentionally shows no bit-diagram; see the target/version requirements below for what governs how this instruction compiles.
PTX ISA Version Introduced PTX ISA 8.0
Minimum Target sm_90

Syntax Forms

One mnemonic covers many type / state-space / scope / modifier combinations - each row below is an independently valid form.

Syntax Data Types State Space(s) Modifiers Min. Target Description
cp.reduce.async.bulk{.sem.scope}.dst.src.completion_mechanism.redOp.type sm_90 cp.reduce.async.bulk is a non-blocking instruction which initiates an asynchronous reduction operation on an array of memory locations specified by the destination address operand dstMem with the sour… (see the official PTX ISA docs for the full description)

Operands

At a Glance

Data Types -

Related

More in Data Movement and Conversion Instructions

Reference

NVIDIA PTX ISA

Description

cp.reduce.async.bulk is a non-blocking instruction which initiates an asynchronous reduction operation on an array of memory locations specified by the destination address operand dstMem with the source array whose location is specified by the source address operand srcMem. The size of the source and the destination array must be the same and is specified by the operand size. Each data element in the destination array is reduced inline with the corresponding data element in the source array with the reduction operation specified by the modifier.redOp. (see the official PTX ISA docs for the full description)

Examples

cp.reduce.async.bulk.shared::cluster.shared::cta.mbarrier::complete_tx::bytes.add.u64
                                                                  [dstMem], [srcMem], size, [mbar];

cp.reduce.async.bulk.shared::cluster.shared::cta.mbarrier::complete_tx::bytes.min.s32
                                                                  [dstMem], [srcMem], size, [mbar];
// (truncated - see the official PTX ISA docs for the full example)

Reproduced from NVIDIA's official PTX ISA documentation for technical accuracy.

Sources