flate: Improve speed in big stateless blocks. #718
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Don't re-alloc and copy dict for every block when compressing more than 32KB. ``` benchmark old ns/op new ns/op delta BenchmarkEncodeDigitsSL1e4-32 52954 52850 -0.20% BenchmarkEncodeDigitsSL1e5-32 781061 745420 -4.56% BenchmarkEncodeDigitsSL1e6-32 8143640 7715674 -5.26% BenchmarkEncodeTwainSL1e4-32 68150 68415 +0.39% BenchmarkEncodeTwainSL1e5-32 715140 687326 -3.89% BenchmarkEncodeTwainSL1e6-32 7718175 7339694 -4.90% benchmark old MB/s new MB/s speedup BenchmarkEncodeDigitsSL1e4-32 188.84 189.21 1.00x BenchmarkEncodeDigitsSL1e5-32 128.03 134.15 1.05x BenchmarkEncodeDigitsSL1e6-32 122.80 129.61 1.06x BenchmarkEncodeTwainSL1e4-32 146.74 146.17 1.00x BenchmarkEncodeTwainSL1e5-32 139.83 145.49 1.04x BenchmarkEncodeTwainSL1e6-32 129.56 136.25 1.05x benchmark old allocs new allocs delta BenchmarkEncodeDigitsSL1e4-32 0 0 +0.00% BenchmarkEncodeDigitsSL1e5-32 3 0 -100.00% BenchmarkEncodeDigitsSL1e6-32 41 0 -100.00% BenchmarkEncodeTwainSL1e4-32 0 0 +0.00% BenchmarkEncodeTwainSL1e5-32 3 0 -100.00% BenchmarkEncodeTwainSL1e6-32 41 0 -100.00% benchmark old bytes new bytes delta BenchmarkEncodeDigitsSL1e4-32 0 0 +0.00% BenchmarkEncodeDigitsSL1e5-32 92929 9 -99.99% BenchmarkEncodeDigitsSL1e6-32 1298964 97 -99.99% BenchmarkEncodeTwainSL1e4-32 0 0 +0.00% BenchmarkEncodeTwainSL1e5-32 92928 8 -99.99% BenchmarkEncodeTwainSL1e6-32 1298871 92 -99.99% ```
kodiakhq bot
referenced
this pull request
in cloudquery/cloudquery
Feb 1, 2023
…7575) This PR contains the following updates: | Package | Type | Update | Change | |---|---|---|---| | [github.com/klauspost/compress](https://togithub.com/klauspost/compress) | indirect | patch | `v1.15.11` -> `v1.15.15` | --- ### Release Notes <details> <summary>klauspost/compress</summary> ### [`v1.15.15`](https://togithub.com/klauspost/compress/releases/tag/v1.15.15) [Compare Source](https://togithub.com/klauspost/compress/compare/v1.15.14...v1.15.15) ##### What's Changed - zstd: Add delta encoding support by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/728](https://togithub.com/klauspost/compress/pull/728) - huff0: Reduce bounds checking by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/734](https://togithub.com/klauspost/compress/pull/734) - huff0: Assembler improvements by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/736](https://togithub.com/klauspost/compress/pull/736) - deflate: Improve level 7-9 by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/739](https://togithub.com/klauspost/compress/pull/739) - gzhttp: Add SuffixETag() and DropETag() options to prevent ETag collisions on compressed responses by [@​willbicks](https://togithub.com/willbicks) in [https://github.com/klauspost/compress/pull/740](https://togithub.com/klauspost/compress/pull/740) - zstd: Don't allocate dataStorage when using byteBuf by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/741](https://togithub.com/klauspost/compress/pull/741) - huff0: Speed up compression of short blocks by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/744](https://togithub.com/klauspost/compress/pull/744) - zstd: Handle dicts by pointer, always by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/743](https://togithub.com/klauspost/compress/pull/743) - fse: Optimize compression by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/745](https://togithub.com/klauspost/compress/pull/745) - Retract v1.14.1-v.1.14.3 by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/742](https://togithub.com/klauspost/compress/pull/742) ##### New Contributors - [@​willbicks](https://togithub.com/willbicks) made their first contribution in [https://github.com/klauspost/compress/pull/740](https://togithub.com/klauspost/compress/pull/740) **Full Changelog**: klauspost/compress@v1.15.14...v1.15.15 ### [`v1.15.14`](https://togithub.com/klauspost/compress/releases/tag/v1.15.14) [Compare Source](https://togithub.com/klauspost/compress/compare/v1.15.13...v1.15.14) #### What's Changed - flate: Improve speed in big stateless blocks. by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/718](https://togithub.com/klauspost/compress/pull/718) - zstd: Trigger BCE by switching on lengths by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/716](https://togithub.com/klauspost/compress/pull/716) - zstd: Shave some instructions off the amd64 asm by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/720](https://togithub.com/klauspost/compress/pull/720) - export NoGzipResponseWriter for custom ResponseWriter wrappers by [@​harshavardhana](https://togithub.com/harshavardhana) in [https://github.com/klauspost/compress/pull/722](https://togithub.com/klauspost/compress/pull/722) - s2: Add example for indexing and existing stream by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/723](https://togithub.com/klauspost/compress/pull/723) - tests: Tweak fuzz tests by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/719](https://togithub.com/klauspost/compress/pull/719) #### New Contributors - [@​harshavardhana](https://togithub.com/harshavardhana) made their first contribution in [https://github.com/klauspost/compress/pull/722](https://togithub.com/klauspost/compress/pull/722) **Full Changelog**: klauspost/compress@v1.15.13...v1.15.14 ### [`v1.15.13`](https://togithub.com/klauspost/compress/releases/tag/v1.15.13) [Compare Source](https://togithub.com/klauspost/compress/compare/v1.15.12...v1.15.13) #### What's Changed - zstd: Add MaxEncodedSize to encoder by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/691](https://togithub.com/klauspost/compress/pull/691) - zstd: Improve "best" end search by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/693](https://togithub.com/klauspost/compress/pull/693) - zstd: Replace bytes.Equal with smaller comparisons by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/695](https://togithub.com/klauspost/compress/pull/695) - zstd: Faster CRC checking/skipping by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/696](https://togithub.com/klauspost/compress/pull/696) - zstd: Rewrite matchLen to make it inlineable by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/701](https://togithub.com/klauspost/compress/pull/701) - zstd: Write table clearing in a way that the compiler recognizes by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/702](https://togithub.com/klauspost/compress/pull/702) - zstd: Use individual reset threshold by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/703](https://togithub.com/klauspost/compress/pull/703) - huff0: Check for zeros earlier in Scratch.countSimple by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/704](https://togithub.com/klauspost/compress/pull/704) - zstd: Improve best compression's match selection by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/705](https://togithub.com/klauspost/compress/pull/705) - zstd: Select best match using selection trees by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/706](https://togithub.com/klauspost/compress/pull/706) - zstd: sync xxhash with final accepted patch upstream by [@​lizthegrey](https://togithub.com/lizthegrey) in [https://github.com/klauspost/compress/pull/707](https://togithub.com/klauspost/compress/pull/707) - zstd: Import xxhash v2.2.0 by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/708](https://togithub.com/klauspost/compress/pull/708) **Full Changelog**: klauspost/compress@v1.15.12...v1.15.13 ### [`v1.15.12`](https://togithub.com/klauspost/compress/releases/tag/v1.15.12) [Compare Source](https://togithub.com/klauspost/compress/compare/v1.15.11...v1.15.12) ##### What's Changed - zstd: Tweak decoder allocs. by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/680](https://togithub.com/klauspost/compress/pull/680) - gzhttp: Always delete `HeaderNoCompression` by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/683](https://togithub.com/klauspost/compress/pull/683) **Full Changelog**: klauspost/compress@v1.15.11...v1.15.12 </details> --- ### Configuration 📅 **Schedule**: Branch creation - "before 3am on the first day of the month" (UTC), Automerge - At any time (no schedule defined). 🚦 **Automerge**: Disabled by config. Please merge this manually once you are satisfied. ♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox. 🔕 **Ignore**: Close this PR and you won't be reminded about this update again. --- - [ ] <!-- rebase-check -->If you want to rebase/retry this PR, check this box --- This PR has been generated by [Renovate Bot](https://togithub.com/renovatebot/renovate). <!--renovate-debug:eyJjcmVhdGVkSW5WZXIiOiIzNC4xMDkuMSIsInVwZGF0ZWRJblZlciI6IjM0LjEwOS4xIn0=-->
kodiakhq bot
referenced
this pull request
in cloudquery/filetypes
Mar 1, 2023
This PR contains the following updates: | Package | Type | Update | Change | |---|---|---|---| | [github.com/klauspost/compress](https://togithub.com/klauspost/compress) | indirect | minor | `v1.15.11` -> `v1.16.0` | --- ### ⚠ Dependency Lookup Warnings ⚠ Warnings were logged while processing this repo. Please check the Dependency Dashboard for more information. --- ### Release Notes <details> <summary>klauspost/compress</summary> ### [`v1.16.0`](https://togithub.com/klauspost/compress/releases/tag/v1.16.0) [Compare Source](https://togithub.com/klauspost/compress/compare/v1.15.15...v1.16.0) #### What's Changed - s2: Add Dictionary support by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/685](https://togithub.com/klauspost/compress/pull/685) - s2: Add Compression Size Estimate by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/752](https://togithub.com/klauspost/compress/pull/752) - s2: Add support for custom stream encoder by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/755](https://togithub.com/klauspost/compress/pull/755) - s2: Add LZ4 block converter by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/748](https://togithub.com/klauspost/compress/pull/748) - s2: Support io.ReaderAt in ReadSeeker by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/747](https://togithub.com/klauspost/compress/pull/747) - s2c/s2sx: Use concurrent decoding by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/746](https://togithub.com/klauspost/compress/pull/746) - tests: Upgrade to Go 1.20 by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/749](https://togithub.com/klauspost/compress/pull/749) - Update all (command) dependencies by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/758](https://togithub.com/klauspost/compress/pull/758) **Full Changelog**: klauspost/compress@v1.15.15...v1.16.0 ### [`v1.15.15`](https://togithub.com/klauspost/compress/releases/tag/v1.15.15) [Compare Source](https://togithub.com/klauspost/compress/compare/v1.15.14...v1.15.15) #### What's Changed - zstd: Add delta encoding support by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/728](https://togithub.com/klauspost/compress/pull/728) - huff0: Reduce bounds checking by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/734](https://togithub.com/klauspost/compress/pull/734) - huff0: Assembler improvements by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/736](https://togithub.com/klauspost/compress/pull/736) - deflate: Improve level 7-9 by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/739](https://togithub.com/klauspost/compress/pull/739) - gzhttp: Add SuffixETag() and DropETag() options to prevent ETag collisions on compressed responses by [@​willbicks](https://togithub.com/willbicks) in [https://github.com/klauspost/compress/pull/740](https://togithub.com/klauspost/compress/pull/740) - zstd: Don't allocate dataStorage when using byteBuf by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/741](https://togithub.com/klauspost/compress/pull/741) - huff0: Speed up compression of short blocks by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/744](https://togithub.com/klauspost/compress/pull/744) - zstd: Handle dicts by pointer, always by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/743](https://togithub.com/klauspost/compress/pull/743) - fse: Optimize compression by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/745](https://togithub.com/klauspost/compress/pull/745) - Retract v1.14.1-v.1.14.3 by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/742](https://togithub.com/klauspost/compress/pull/742) #### New Contributors - [@​willbicks](https://togithub.com/willbicks) made their first contribution in [https://github.com/klauspost/compress/pull/740](https://togithub.com/klauspost/compress/pull/740) **Full Changelog**: klauspost/compress@v1.15.14...v1.15.15 ### [`v1.15.14`](https://togithub.com/klauspost/compress/releases/tag/v1.15.14) [Compare Source](https://togithub.com/klauspost/compress/compare/v1.15.13...v1.15.14) #### What's Changed - flate: Improve speed in big stateless blocks. by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/718](https://togithub.com/klauspost/compress/pull/718) - zstd: Trigger BCE by switching on lengths by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/716](https://togithub.com/klauspost/compress/pull/716) - zstd: Shave some instructions off the amd64 asm by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/720](https://togithub.com/klauspost/compress/pull/720) - export NoGzipResponseWriter for custom ResponseWriter wrappers by [@​harshavardhana](https://togithub.com/harshavardhana) in [https://github.com/klauspost/compress/pull/722](https://togithub.com/klauspost/compress/pull/722) - s2: Add example for indexing and existing stream by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/723](https://togithub.com/klauspost/compress/pull/723) - tests: Tweak fuzz tests by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/719](https://togithub.com/klauspost/compress/pull/719) #### New Contributors - [@​harshavardhana](https://togithub.com/harshavardhana) made their first contribution in [https://github.com/klauspost/compress/pull/722](https://togithub.com/klauspost/compress/pull/722) **Full Changelog**: klauspost/compress@v1.15.13...v1.15.14 ### [`v1.15.13`](https://togithub.com/klauspost/compress/releases/tag/v1.15.13) [Compare Source](https://togithub.com/klauspost/compress/compare/v1.15.12...v1.15.13) #### What's Changed - zstd: Add MaxEncodedSize to encoder by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/691](https://togithub.com/klauspost/compress/pull/691) - zstd: Improve "best" end search by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/693](https://togithub.com/klauspost/compress/pull/693) - zstd: Replace bytes.Equal with smaller comparisons by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/695](https://togithub.com/klauspost/compress/pull/695) - zstd: Faster CRC checking/skipping by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/696](https://togithub.com/klauspost/compress/pull/696) - zstd: Rewrite matchLen to make it inlineable by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/701](https://togithub.com/klauspost/compress/pull/701) - zstd: Write table clearing in a way that the compiler recognizes by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/702](https://togithub.com/klauspost/compress/pull/702) - zstd: Use individual reset threshold by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/703](https://togithub.com/klauspost/compress/pull/703) - huff0: Check for zeros earlier in Scratch.countSimple by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/704](https://togithub.com/klauspost/compress/pull/704) - zstd: Improve best compression's match selection by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/705](https://togithub.com/klauspost/compress/pull/705) - zstd: Select best match using selection trees by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/706](https://togithub.com/klauspost/compress/pull/706) - zstd: sync xxhash with final accepted patch upstream by [@​lizthegrey](https://togithub.com/lizthegrey) in [https://github.com/klauspost/compress/pull/707](https://togithub.com/klauspost/compress/pull/707) - zstd: Import xxhash v2.2.0 by [@​greatroar](https://togithub.com/greatroar) in [https://github.com/klauspost/compress/pull/708](https://togithub.com/klauspost/compress/pull/708) **Full Changelog**: klauspost/compress@v1.15.12...v1.15.13 ### [`v1.15.12`](https://togithub.com/klauspost/compress/releases/tag/v1.15.12) [Compare Source](https://togithub.com/klauspost/compress/compare/v1.15.11...v1.15.12) #### What's Changed - zstd: Tweak decoder allocs. by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/680](https://togithub.com/klauspost/compress/pull/680) - gzhttp: Always delete `HeaderNoCompression` by [@​klauspost](https://togithub.com/klauspost) in [https://github.com/klauspost/compress/pull/683](https://togithub.com/klauspost/compress/pull/683) **Full Changelog**: klauspost/compress@v1.15.11...v1.15.12 </details> --- ### Configuration 📅 **Schedule**: Branch creation - "before 3am on the first day of the month" (UTC), Automerge - At any time (no schedule defined). 🚦 **Automerge**: Disabled by config. Please merge this manually once you are satisfied. ♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox. 🔕 **Ignore**: Close this PR and you won't be reminded about this update again. --- - [ ] <!-- rebase-check -->If you want to rebase/retry this PR, check this box --- This PR has been generated by [Renovate Bot](https://togithub.com/renovatebot/renovate). <!--renovate-debug:eyJjcmVhdGVkSW5WZXIiOiIzNC4xMDkuMSIsInVwZGF0ZWRJblZlciI6IjM0LjE1NC4wIn0=-->
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Don't re-alloc and copy dict for every block when compressing more than 32KB.