peermem

Documentation

Overview

Package peermem monitors the peermem module status. Optional, enabled if the host has NVIDIA GPUs.

Index

• Constants
• func New(ctx context.Context, cfg Config) components.Component
• type Config
  • func ParseConfig(b any, db *sql.DB) (*Config, error)
  • func (cfg Config) Validate() error
• type Output
  • func ParseOutputJSON(data []byte) (*Output, error)
  • func ParseStateLsmodPeermem(m map[string]string) (*Output, error)
  • func ParseStatesToOutput(states ...components.State) (*Output, error)
  • func ToOutput(i *nvidia_query.Output) *Output
  • func (o *Output) JSON() ([]byte, error)
  • func (o *Output) States() ([]components.State, error)

Constants

const (
	// repeated messages may indicate more persistent issue on the inter-GPU communication
	// e.g.,
	// [Thu Sep 19 02:29:46 2024] nvidia-peermem nv_get_p2p_free_callback:127 ERROR detected invalid context, skipping further processing
	// [Thu Sep 19 02:29:46 2024] nvidia-peermem nv_get_p2p_free_callback:127 ERROR detected invalid context, skipping further processing
	// [Thu Sep 19 02:29:46 2024] nvidia-peermem nv_get_p2p_free_callback:127 ERROR detected invalid context, skipping further processing
	EventNamePeermemInvalidContextFromDmesg = "peermem_invalid_context_from_dmesg"

	EventKeyPeermemInvalidContextFromDmesgUnixSeconds = "unix_seconds"
	EventKeyPeermemInvalidContextFromDmesgLogLine     = "log_line"
)

const (
	StateNameLsmodPeermem = "lsmod_peermem"

	StateKeyLsmodPeermemData           = "data"
	StateKeyLsmodPeermemEncoding       = "encoding"
	StateValueLsmodPeermemEncodingJSON = "json"
)

Functions

func New

func New(ctx context.Context, cfg Config) components.Component

Types

type Config

type Config struct {
	Query query_config.Config `json:"query"`
}

func ParseConfig

func ParseConfig(b any, db *sql.DB) (*Config, error)

func (Config) Validate

func (cfg Config) Validate() error

type Output

type Output struct {
	// Represents the number of GPUs in the system.
	// This is used to determine if ibcore may be expected to use peermem module.
	GPUCount int `json:"gpu_count"`

	ProductName  string                           `json:"product_name"`
	LsmodPeermem peermem.LsmodPeermemModuleOutput `json:"lsmod_peermem"`
}

func ParseOutputJSON

func ParseOutputJSON(data []byte) (*Output, error)

func ParseStateLsmodPeermem

func ParseStateLsmodPeermem(m map[string]string) (*Output, error)

func ParseStatesToOutput

func ParseStatesToOutput(states ...components.State) (*Output, error)

func ToOutput

func ToOutput(i *nvidia_query.Output) *Output

ToOutput converts nvidia_query.Output to Output. It returns an empty non-nil object, if the input or the required field is nil (e.g., i.SMI).

func (*Output) JSON

func (o *Output) JSON() ([]byte, error)

func (*Output) States

func (o *Output) States() ([]components.State, error)