pprof

Documentation

Index

• func FromBytes(input []byte, fn func(*profilev1.Profile, int) error) error
• func FromProfile(p *profile.Profile) (*profilev1.Profile, error)
• type Profile
  • func OpenFile(path string) (*Profile, error)
  • func RawFromBytes(input []byte) (_ *Profile, err error)
  • func (p *Profile) Close()
  • func (p *Profile) Normalize()
  • func (p *Profile) SizeBytes() int
  • func (p *Profile) WriteTo(w io.Writer) (int64, error)
• type StacktracesHasher
  • func (h StacktracesHasher) Hashes(samples []*profilev1.Sample) []uint64

Functions

func FromBytes

func FromBytes(input []byte, fn func(*profilev1.Profile, int) error) error

Read Profile from Bytes

func FromProfile

func FromProfile(p *profile.Profile) (*profilev1.Profile, error)

Types

type Profile

type Profile struct {
	*profilev1.Profile
	// contains filtered or unexported fields
}

func OpenFile

func OpenFile(path string) (*Profile, error)

func RawFromBytes

func RawFromBytes(input []byte) (_ *Profile, err error)

Read RawProfile from bytes

func (*Profile) Close

func (p *Profile) Close()

func (*Profile) Normalize

func (p *Profile) Normalize()

Normalize normalizes the profile by:

• Removing all duplicate samples (summing their values).
• Removing redundant profile labels (byte => unique of an allocation site) todo: We should reassess if this was a good choice because by merging duplicate stacktrace samples we cannot recompute the allocation per site ("bytes") profile label.
• Removing empty samples.
• Then remove unused references.
• Ensure that the profile has a time_nanos set
• Removes addresses from symbolized profiles.

func (*Profile) SizeBytes

func (p *Profile) SizeBytes() int

func (*Profile) WriteTo

func (p *Profile) WriteTo(w io.Writer) (int64, error)

WriteTo writes the profile to the given writer.

type StacktracesHasher

type StacktracesHasher struct {
	// contains filtered or unexported fields
}

func (StacktracesHasher) Hashes

func (h StacktracesHasher) Hashes(samples []*profilev1.Sample) []uint64

todo we might want to reuse the results to avoid allocations