fabric.try_pullred GPU Virtual ISA NVIDIA

fabric.try_pullred Fabric Instructions

fabric.try_pullred.async.multimem.dst.completion_mechanism.sem.scope.redOpBit.typeBit.sync [dst], [srcLeId, srcDataOff], size, [bar], imm-membermask;

Initiates asynchronous loads from multiple resources pointed to by multicast fabric handle [srcLeId, srcDataOff], of size bytes, and performs element-

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 9.3
Minimum Target sm_100

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
fabric.try_pullred.async.multimem.dst.completion_mechanism.sem.scope.redOpBit.typeBit.sync [dst], [srcLeId, srcDataOff], size, [bar], imm-membermask; sm_100 Initiates asynchronous loads from multiple resources pointed to by multicast fabric handle [srcLeId, srcDataOff], of size bytes, and performs element-wise reduction on data across each of the loads. (see the official PTX ISA docs for the full description)

Operands

At a Glance

Data Types -

Related

More in Fabric Instructions

Reference

NVIDIA PTX ISA

Description

Initiates asynchronous loads from multiple resources pointed to by multicast fabric handle [srcLeId, srcDataOff], of size bytes, and performs element-wise reduction on data across each of the loads. The result is stored in [dst, dst + size - 1]. srcLeId is a 32-bit unsigned value denoting the logical endpoint identifier, and srcDataOff is a 64-bit unsigned value denoting the base offset of the resources to access within the multicast logical endpoint associated with srcLeId. The size operand is 32 bits and specifies the number of bytes to be copied. It must be a multiple of 16; otherwise, the behavior is undefined. (see the official PTX ISA docs for the full description)

Examples

fabric.try_pullred.async.multimem.mbarrier::complete_tx::bytes.mbarrier::report::fabric.shared::cta.f32.add.sync.relaxed.sys [dst], [leId, offset], size, [mbar], imm-membermask;

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

Sources