tensormap.replace GPU Virtual ISA NVIDIA

tensormap.replace Data Movement and Conversion Instructions

tensormap.replace.mode.field1{.ss}.b1024.type [addr], new_val;

The tensormap.replace instruction replaces the field, specified by.field qualifier, of the tensor-map object at the location specified by the address operand addr with a new value.

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 8.3
Minimum Target sm_90a

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
tensormap.replace.mode.field1{.ss}.b1024.type [addr], new_val; sm_90a The tensormap.replace instruction replaces the field, specified by.field qualifier, of the tensor-map object at the location specified by the address operand addr with a new value. (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

The tensormap.replace instruction replaces the field, specified by.field qualifier, of the tensor-map object at the location specified by the address operand addr with a new value. The new value is specified by the argument new_val. Qualifier.mode specifies the mode of the tensor-map object located at the address operand addr. Instruction type.b1024 indicates the size of the tensor-map object, which is 1024 bits. Operand new_val has the type.type. When.field is specified as.global_address or.global_stride,.type must be.b64. (see the official PTX ISA docs for the full description)

Examples

tensormap.replace.tile.global_address.shared::cta.b1024.b64   [sMem], new_val;

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

Sources