tcgen05.commit GPU Virtual ISA NVIDIA

tcgen05.commit TensorCore 5th Generation Family Instructions

tcgen05.commit.cta_group.completion_mechanism{.shared::cluster}{.multicast}.b64

The instruction tcgen05.commit is an asynchronous instruction which makes the mbarrier object, specified by the address operand mbar, track the completion of all the prior asynchronous tcgen05 operations, as listed in mbarrier based completion mechanism, initiated by the executing thread.

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 8.6
Minimum Target sm_100a

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
tcgen05.commit.cta_group.completion_mechanism{.shared::cluster}{.multicast}.b64 sm_100a The instruction tcgen05.commit is an asynchronous instruction which makes the mbarrier object, specified by the address operand mbar, track the completion of all the prior asynchronous tcgen05 operati… (see the official PTX ISA docs for the full description)

Operands

At a Glance

Data Types -

Reference

NVIDIA PTX ISA

Description

The instruction tcgen05.commit is an asynchronous instruction which makes the mbarrier object, specified by the address operand mbar, track the completion of all the prior asynchronous tcgen05 operations, as listed in mbarrier based completion mechanism, initiated by the executing thread. Upon the completion of the tracked asynchronous tcgen05 operations, the signal specified by the.completion_mechanism is triggered by the system on the mbarrier object. This instruction accesses its mbarrier operand using generic-proxy. The instruction tcgen05.commit.cta_group::1 tracks for the completion of all prior asynchronous tcgen05 operations with.cta_group::1 issued by the current thread. (see the official PTX ISA docs for the full description)

Examples

Example 1:
tcgen05.cp.cta_group::1.128x256b                      [taddr0], sdesc0;
tcgen05.commit.cta_group::1.mbarrier::arrive::one.b64 [mbarObj1];

loop:
mbarrier.try_wait.parity.b64 p, [mbarObj1], 0;
// (truncated - see the official PTX ISA docs for the full example)

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

Sources