Yann Collet
44cd3c8451
disable asyncio during compression
...
it's not helpful for performance, and can in some times be detrimental.
the synchronous mode is also much easier to follow and debug.
2025-10-25 00:50:55 -07:00
Yann Collet and GitHub
6e58e9a2c4
Merge pull request #4509 from arpadpanyik-arm/revert_branch_opt
...
AArch64: Revert previous branch optimization
2025-10-18 10:10:55 -07:00
Arpad Panyik
0dffae42e3
AArch64: Remove 32-bit code from ZSTD_decodeSequence
...
Remove the 32-bit code paths from the AArch64 only sections of
ZSTD_decodeSequence.
2025-10-08 18:59:24 +00:00
Arpad Panyik
33618c89e5
AArch64: Revert previous branch optimization
...
Revert a branch optimization that was based on an incorrect
assumption in the AArch64 part of ZSTD_decodeSequence. In extreme
cases the existing implementation could lead to data corruption.
Insert an UNLIKELY hint to guide the compilers toward generating more
efficient machine code.
2025-10-08 18:58:45 +00:00
Yann Collet and GitHub
23dae4bf49
Merge pull request #4504 from rzikm/asm-conditional
...
Make ASM language conditional at CMake configure time
2025-10-06 23:39:28 -07:00
Radek Zikmund
f94a01de74
Enable ASM language during CMake configure time only when needed
2025-10-02 14:40:11 +02:00
Yann Collet and GitHub
b8d6101fba
Merge pull request #4502 from zijianli1234/dev
...
- Modify the GCC version used for CI testing of the RISCV architecture
2025-09-30 12:35:40 -07:00
ZijianLi
87cc127705
- Modify the GCC version used for CI testing of the RISCV architecture
...
- Fix a bug in the ZSTD_row_getRVVMask function
- Improve some performance for ZSTD_copy16()
2025-09-26 22:34:57 +08:00
Yann Collet and GitHub
f42dee27b6
Merge pull request #4500 from facebook/legacy01_init
...
fix minor initialization warnings
2025-09-24 22:46:05 -07:00
Yann Collet
17888b3fbe
fix minor initialization warnings
2025-09-24 22:08:03 -07:00
Yann Collet and GitHub
a8f732e50f
Merge pull request #4498 from facebook/contentSize_doc
...
update documentation of ZSTD_getFrameContentSize()
2025-09-23 23:59:34 -07:00
Yann Collet
c15fa3cd40
update documentation of ZSTD_getFrameContentSize()
...
hopefully answering #4495
2025-09-23 23:17:11 -07:00
Yann Collet and GitHub
89d685e42d
Merge pull request #4497 from facebook/legacy_fix
...
fix minor warning in legacy decoders
2025-09-23 12:55:50 -08:00
Yann Collet
4c1f86c777
fix minor warning in legacy decoders
...
for mingw + clang CI test
2025-09-23 13:01:38 -07:00
Yann Collet and GitHub
ee9b5f4371
Merge pull request #4490 from facebook/doc_profiling_vs
...
minor: add profiling notes for Visual Studio
2025-09-20 22:30:40 -08:00
Yann Collet
0c6d32df9a
added profiling notes for Visual Studio
2025-09-20 22:13:49 -07:00
Yann Collet and GitHub
f9b0c4fd0d
Merge pull request #4489 from facebook/doc_install
...
Added documentation details for Makefile installation and pkg-config.
2025-09-20 21:04:35 -08:00
Yann Collet
be072c708e
Added documentation details for Makefile installation and pkg-config.
2025-09-20 16:33:41 +00:00
Yann Collet and GitHub
085cc9319a
Merge pull request #4486 from rlefko/fix-pthread-init-memleak
...
Fix memory leak in pthread init functions on failure
2025-09-19 21:42:21 -08:00
Yann Collet and GitHub
e6e5a95c15
Merge pull request #4487 from neiljohari/adhoc/dictionary-file-counting
...
make DiB_fileStats skip invalid files (fileSize <= 0) to prevent negative totals and bogus allocation
2025-09-19 21:42:06 -08:00
Yann Collet and GitHub
d190b38863
Merge pull request #4481 from w1m024/support-rvv-getmask
...
add RVV optimization for ZSTD_row_getMatchMask
2025-09-16 13:55:11 -08:00
Neil Johari
236e44f00f
Remove debug logging
2025-09-16 00:03:08 -07:00
Neil Johari
85f4a7e84e
Fix bug
2025-09-16 00:02:04 -07:00
Neil Johari
96fdb9bd16
Add debug logging and simple repro
2025-09-15 23:58:45 -07:00
Ryan Lefkowitz
c59812e558
🔧 Fix memory leak in pthread init functions on failure
...
When pthread_mutex_init() or pthread_cond_init() fails in the debug
implementation (DEBUGLEVEL >= 1), the previously allocated memory was
not freed, causing a memory leak.
This fix ensures that allocated memory is properly freed when pthread
initialization functions fail, preventing resource leaks in error
conditions.
The issue affects:
- ZSTD_pthread_mutex_init() at lib/common/threading.c:146
- ZSTD_pthread_cond_init() at lib/common/threading.c:167
This is particularly important for long-running applications or
scenarios with resource constraints where pthread initialization
might fail due to system limits.
2025-09-15 18:20:01 -04:00
w1m024 and gong-flying
fb7a86f20f
Refactor ZSTD_row_getMatchMask for RVV optimization
...
Performance (vs. SWAR)
- 16-byte data: 5.87x speedup
- 32-byte data: 9.63x speedup
- 64-byte data: 17.98x speedup
Co-authored-by: gong-flying <gongxiaofei24@iscas.ac.cn >
2025-09-11 20:45:54 +00:00
w1m024 and gong-flying
c9d2cbd5ba
add RVV optimization for ZSTD_row_getMatchMask
...
Co-authored-by: gong-flying <gongxiaofei24@iscas.ac.cn >
2025-09-09 06:20:55 +00:00
Yann Collet and GitHub
98d2b90e82
Merge pull request #4480 from facebook/dependabot/github_actions/github/codeql-action-3.30.1
...
Bump github/codeql-action from 3.29.4 to 3.30.1
2025-09-08 07:52:31 -07:00
Yann Collet and GitHub
683c91fca8
Merge pull request #4479 from facebook/dependabot/github_actions/msys2/setup-msys2-2.29.0
...
Bump msys2/setup-msys2 from 2.28.0 to 2.29.0
2025-09-08 07:51:17 -07:00
dependabot[bot] and GitHub
ef4bab9079
Bump github/codeql-action from 3.29.4 to 3.30.1
...
Bumps [github/codeql-action](https://github.com/github/codeql-action ) from 3.29.4 to 3.30.1.
- [Release notes](https://github.com/github/codeql-action/releases )
- [Changelog](https://github.com/github/codeql-action/blob/main/CHANGELOG.md )
- [Commits](https://github.com/github/codeql-action/compare/4e828ff8d448a8a6e532957b1811f387a63867e8...f1f6e5f6af878fb37288ce1c627459e94dbf7d01 )
---
updated-dependencies:
- dependency-name: github/codeql-action
dependency-version: 3.30.1
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com >
2025-09-08 05:06:48 +00:00
dependabot[bot] and GitHub
d3536643a9
Bump msys2/setup-msys2 from 2.28.0 to 2.29.0
...
Bumps [msys2/setup-msys2](https://github.com/msys2/setup-msys2 ) from 2.28.0 to 2.29.0.
- [Release notes](https://github.com/msys2/setup-msys2/releases )
- [Changelog](https://github.com/msys2/setup-msys2/blob/main/CHANGELOG.md )
- [Commits](https://github.com/msys2/setup-msys2/compare/40677d36a502eb2cf0fb808cc9dec31bf6152638...fb197b72ce45fb24f17bf3f807a388985654d1f2 )
---
updated-dependencies:
- dependency-name: msys2/setup-msys2
dependency-version: 2.29.0
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com >
2025-09-08 05:06:40 +00:00
Yann Collet and GitHub
3580aa28e6
Merge pull request #4472 from bgilbert/override_dependency
...
meson: Call `meson.override_dependency()` if Meson is new enough
2025-09-05 15:32:16 -07:00
Yann Collet and GitHub
6936bc84e7
Merge pull request #4475 from Cyan4973/default_nbThreads
...
Default nb threads
2025-09-02 17:19:17 -07:00
Yann Collet
249e4a07d5
fixed minor unused variable warning
...
in certain compilation modes
2025-09-02 16:36:44 -07:00
Yann Collet
725a152c7b
benchmark uses 1 thread by default
2025-09-02 16:29:14 -07:00
Yann Collet
c41fc1aa01
only display nbThread Msg in nbThreads > 1
2025-09-02 16:05:35 -07:00
Yann Collet
6551db3459
specify nb of threads used during benchmarking
...
used to require `-v` (verbose) modifier
2025-09-02 15:53:45 -07:00
Yann Collet
1c5d45fc11
fixed -T# documentation in zstd -H
...
provide the local value for default nbThreads
which is dynamic and depends on local nb of cores.
2025-09-02 15:46:51 -07:00
Yann Collet and GitHub
fc551317fb
Merge pull request #4474 from jlokier/threads-doc-fix
...
Update manual about the default value of `-T#`/`--threads=#`
2025-09-02 15:40:32 -07:00
Jamie Lokier
64c4288e92
Update manual about the default value of -T#/--threads=#
...
The section about `ZSTD_NBTHREADS` already explains the default number of
threads, since it changed from 1 (commit 17beeb5 ). But the option description
for `-T#`/`--threads=#` incorrectly said the default was still 1.
I noticed this when I found compression slower with `-T1` than without it.
2025-09-02 16:44:39 +01:00
Benjamin Gilbert
ba59aeb674
meson: Call meson.override_dependency() if Meson is new enough
...
This tells Meson that we intend libzstd_dep to be used by a parent project
if the parent looks for a dependency named "libzstd". Without this, the
mapping from "libzstd" to our variable libzstd_dep must be encoded in the
Meson wrap file or in the parent's meson.build.
2025-08-28 18:50:34 -05:00
Yann Collet and GitHub
22b4483163
Merge pull request #4469 from facebook/dependabot/github_actions/actions/setup-java-5.0.0
...
Bump actions/setup-java from 4.7.1 to 5.0.0
2025-08-25 09:07:01 -07:00
dependabot[bot] and GitHub
0ee187c540
Bump actions/setup-java from 4.7.1 to 5.0.0
...
Bumps [actions/setup-java](https://github.com/actions/setup-java ) from 4.7.1 to 5.0.0.
- [Release notes](https://github.com/actions/setup-java/releases )
- [Commits](https://github.com/actions/setup-java/compare/c5195efecf7bdfc987ee8bae7a71cb8b11521c00...dded0888837ed1f317902acf8a20df0ad188d165 )
---
updated-dependencies:
- dependency-name: actions/setup-java
dependency-version: 5.0.0
dependency-type: direct:production
update-type: version-update:semver-major
...
Signed-off-by: dependabot[bot] <support@github.com >
2025-08-25 09:00:57 +00:00
Yann Collet and GitHub
b5c294ea01
Merge pull request #4440 from arpadpanyik-arm/convert_seq_sve2
...
AArch64: Add SVE2 path for convertSequences_noRepcodes
2025-08-21 17:20:33 -07:00
Arpad Panyik
2849f3a5d1
AArch64: Add SVE2 path for convertSequences_noRepcodes
...
Add an 8-way vector length agnostic (VLA) SVE2 code path for
convertSequences_noRepcodes. It works with any SVE vector length.
Relative performance to GCC-13 using: `./fullbench -b18 -l5 enwik5`
Neon SVE2
Neoverse-V2 before after uplift
GCC-13: 100.000% 103.209% 1.032x
GCC-14: 100.309% 134.872% 1.344x
GCC-15: 100.355% 134.827% 1.343x
Clang-18: 123.614% 128.565% 1.040x
Clang-19: 123.587% 132.984% 1.076x
Clang-20: 123.629% 133.023% 1.075x
Neon SVE2
Cortex-A720 before after uplift
GCC-13: 100.000% 116.032% 1.160x
GCC-14: 99.700% 116.648% 1.169x
GCC-15: 100.354% 117.047% 1.166x
Clang-18: 100.447% 116.762% 1.162x
Clang-19: 100.454% 116.627% 1.160x
Clang-20: 100.452% 116.649% 1.161x
2025-08-21 17:37:41 +00:00
Yann Collet and GitHub
290e692ef8
Merge pull request #4463 from brad0/gnu_source_qsort
...
Check for build environment instead of just _GNU_SOURCE
2025-08-21 09:30:29 -07:00
Yann Collet and GitHub
b3f134bfd4
Merge pull request #4465 from thiru-mcw/arm64_support
...
WOA_support:: Add CI setup for packaging Windows on ARM artifacts
2025-08-20 11:23:34 -07:00
Thirumalai Nagalingam
076283fa88
CI: Enable MSVC ARM64 job using Github WOA runner
...
- Reintroduce the MSVC ARM64 build configuration with "Visual Studio 17 2022"
- Update runner to `windows-11-arm` (GitHub-hosted Windows on ARM)
2025-08-20 17:19:48 +05:30
Thirumalai Nagalingam
002c227596
CI: Add CI setup for packaging Win-ARM64 artifacts
2025-08-20 17:12:21 +05:30
Thirumalai Nagalingam
42243c3d46
CI: Update build_package.bat for CMake builds
2025-08-20 17:12:05 +05:30