Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

zstd: Use single stream literal encoding #146

Merged
merged 1 commit into from
Aug 4, 2019
Merged

zstd: Use single stream literal encoding #146

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
60 changes: 60 additions & 0 deletions huff0/compress_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -519,3 +519,63 @@ func BenchmarkCompress4XReusePrefer(b *testing.B) {
})
}
}

func BenchmarkCompress1XSizes(b *testing.B) {
test := testfiles[0]
sizes := []int{1e2, 2e2, 5e2, 1e3, 5e3, 1e4, 5e4}
for _, size := range sizes {
b.Run(test.name+"-"+fmt.Sprint(size), func(b *testing.B) {
var s Scratch
s.Reuse = ReusePolicyNone
buf0, err := test.fn()
if err != nil {
b.Fatal(err)
}
buf0 = buf0[:size]
_, re, err := Compress1X(buf0, &s)
if err != test.err1X {
b.Fatal("unexpected error:", err)
}
//b.Log("Size:", len(o))
b.ResetTimer()
b.ReportAllocs()
b.SetBytes(int64(len(buf0)))
for i := 0; i < b.N; i++ {
_, re, _ = Compress1X(buf0, &s)
if re {
b.Fatal("reused")
}
}
})
}
}

func BenchmarkCompress4XSizes(b *testing.B) {
test := testfiles[0]
sizes := []int{1e2, 2e2, 5e2, 1e3, 5e3, 1e4, 5e4}
for _, size := range sizes {
b.Run(test.name+"-"+fmt.Sprint(size), func(b *testing.B) {
var s Scratch
s.Reuse = ReusePolicyNone
buf0, err := test.fn()
if err != nil {
b.Fatal(err)
}
buf0 = buf0[:size]
_, re, err := Compress4X(buf0, &s)
if err != test.err1X {
b.Fatal("unexpected error:", err)
}
//b.Log("Size:", len(o))
b.ResetTimer()
b.ReportAllocs()
b.SetBytes(int64(len(buf0)))
for i := 0; i < b.N; i++ {
_, re, _ = Compress4X(buf0, &s)
if re {
b.Fatal("reused")
}
}
})
}
}
60 changes: 47 additions & 13 deletions zstd/blockenc.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -155,14 +155,17 @@ func (h *literalsHeader) setSize(regenLen int) {
}

// setSizes will set the size of a compressed literals section and the input length.
func (h *literalsHeader) setSizes(compLen, inLen int) {
func (h *literalsHeader) setSizes(compLen, inLen int, single bool) {
compBits, inBits := bits.Len32(uint32(compLen)), bits.Len32(uint32(inLen))
// Only retain 2 bits
const mask = 3
lh := uint64(*h & mask)
switch {
case compBits <= 10 && inBits <= 10:
lh |= (1 << 2) | (uint64(inLen) << 4) | (uint64(compLen) << (10 + 4)) | (3 << 60)
if !single {
lh |= 1 << 2
}
lh |= (uint64(inLen) << 4) | (uint64(compLen) << (10 + 4)) | (3 << 60)
if debug {
const mmask = (1 << 24) - 1
n := (lh >> 4) & mmask
Expand All @@ -175,8 +178,14 @@ func (h *literalsHeader) setSizes(compLen, inLen int) {
}
case compBits <= 14 && inBits <= 14:
lh |= (2 << 2) | (uint64(inLen) << 4) | (uint64(compLen) << (14 + 4)) | (4 << 60)
if single {
panic("single stream used with more than 10 bits length.")
}
case compBits <= 18 && inBits <= 18:
lh |= (3 << 2) | (uint64(inLen) << 4) | (uint64(compLen) << (18 + 4)) | (5 << 60)
if single {
panic("single stream used with more than 10 bits length.")
}
default:
panic("internal error: block too big")
}
Expand Down Expand Up @@ -307,12 +316,30 @@ func (b *blockEnc) encodeLits() error {
return nil
}

// TODO: Switch to 1X when less than x bytes.
out, reUsed, err := huff0.Compress4X(b.literals, b.litEnc)
// Bail out of compression is too little.
if len(out) > (len(b.literals) - len(b.literals)>>4) {
var (
out []byte
reUsed, single bool
err error
)
if len(b.literals) >= 1024 {
// Use 4 Streams.
out, reUsed, err = huff0.Compress4X(b.literals, b.litEnc)
if len(out) > len(b.literals)-len(b.literals)>>4 {
// Bail out of compression is too little.
err = huff0.ErrIncompressible
}
} else if len(b.literals) > 32 {
// Use 1 stream
single = true
out, reUsed, err = huff0.Compress1X(b.literals, b.litEnc)
if len(out) > len(b.literals)-len(b.literals)>>4 {
// Bail out of compression is too little.
err = huff0.ErrIncompressible
}
} else {
err = huff0.ErrIncompressible
}

switch err {
case huff0.ErrIncompressible:
if debug {
Expand Down Expand Up @@ -351,7 +378,7 @@ func (b *blockEnc) encodeLits() error {
lh.setType(literalsBlockCompressed)
}
// Set sizes
lh.setSizes(len(out), len(b.literals))
lh.setSizes(len(out), len(b.literals), single)
bh.setSize(uint32(len(out) + lh.size() + 1))

// Write block headers.
Expand Down Expand Up @@ -381,16 +408,23 @@ func (b *blockEnc) encode() error {
b.output = bh.appendTo(b.output)

var (
out []byte
reUsed bool
err error
out []byte
reUsed, single bool
err error
)
if len(b.literals) > 32 {
// TODO: Switch to 1X on small blocks.
if len(b.literals) >= 1024 {
// Use 4 Streams.
out, reUsed, err = huff0.Compress4X(b.literals, b.litEnc)
if len(out) > len(b.literals)-len(b.literals)>>4 {
err = huff0.ErrIncompressible
}
} else if len(b.literals) > 32 {
// Use 1 stream
single = true
out, reUsed, err = huff0.Compress1X(b.literals, b.litEnc)
if len(out) > len(b.literals)-len(b.literals)>>4 {
err = huff0.ErrIncompressible
}
} else {
err = huff0.ErrIncompressible
}
Expand Down Expand Up @@ -435,7 +469,7 @@ func (b *blockEnc) encode() error {
}
}
}
lh.setSizes(len(out), len(b.literals))
lh.setSizes(len(out), len(b.literals), single)
if debug {
printf("Compressed %d literals to %d bytes", len(b.literals), len(out))
println("Adding literal header:", lh)
Expand Down