clara/bluenoise-raytracer
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

cleanup and refactoring

Browse source
This commit is contained in:
CDaut 2024-05-25 11:53:25 +02:00
parent 2302158928
commit 76f6bf62a4
Signed by: clara
GPG key ID: 223391B52FAD4463
1285 changed files with 757994 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

800
raytracer/nvpro_core/third_party/zstd/CHANGELOG vendored Normal file
View file

@ -0,0 +1,800 @@
v1.5.5 (Apr 2023)
fix: fix rare corruption bug affecting the high compression mode, reported by @danlark1 (#3517, @terrelln)
perf: improve mid-level compression speed (#3529, #3533, #3543, @yoniko and #3552, @terrelln)
lib: deprecated bufferless block-level API (#3534) by @terrelln
cli: mmap large dictionaries to save memory, by @daniellerozenblit
cli: improve speed of --patch-from mode (~+50%) (#3545) by @daniellerozenblit
cli: improve i/o speed (~+10%) when processing lots of small files (#3479) by @felixhandte
cli: zstd no longer crashes when requested to write into write-protected directory (#3541) by @felixhandte
cli: fix decompression into block device using -o, reported by @georgmu (#3583)
build: fix zstd CLI compiled with lzma support but not zlib support (#3494) by @Hello71
build: fix cmake does no longer require 3.18 as minimum version (#3510) by @kou
build: fix MSVC+ClangCL linking issue (#3569) by @tru
build: fix zstd-dll, version of zstd CLI that links to the dynamic library (#3496) by @yoniko
build: fix MSVC warnings (#3495) by @embg
doc: updated zstd specification to clarify corner cases, by @Cyan4973
doc: document how to create fat binaries for macos (#3568) by @rickmark
misc: improve seekable format ingestion speed (~+100%) for very small chunk sizes (#3544) by @Cyan4973
misc: tests/fullbench can benchmark multiple files (#3516) by @dloidolt
v1.5.4 (Feb 2023)
perf: +20% faster huffman decompression for targets that can't compile x64 assembly (#3449, @terrelln)
perf: up to +10% faster streaming compression at levels 1-2 (#3114, @embg)
perf: +4-13% for levels 5-12 by optimizing function generation (#3295, @terrelln)
pref: +3-11% compression speed for `arm` target (#3199, #3164, #3145, #3141, #3138, @JunHe77 and #3139, #3160, @danlark1)
perf: +5-30% faster dictionary compression at levels 1-4 (#3086, #3114, #3152, @embg)
perf: +10-20% cold dict compression speed by prefetching CDict tables (#3177, @embg)
perf: +1% faster compression by removing a branch in ZSTD_fast_noDict (#3129, @felixhandte)
perf: Small compression ratio improvements in high compression mode (#2983, #3391, @Cyan4973 and #3285, #3302, @daniellerozenblit)
perf: small speed improvement by better detecting `STATIC_BMI2` for `clang` (#3080, @TocarIP)
perf: Improved streaming performance when `ZSTD_c_stableInBuffer` is set (#2974, @Cyan4973)
cli: Asynchronous I/O for improved cli speed (#2975, #2985, #3021, #3022, @yoniko)
cli: Change `zstdless` behavior to align with `zless` (#2909, @binhdvo)
cli: Keep original file if `-c` or `--stdout` is given (#3052, @dirkmueller)
cli: Keep original files when result is concatenated into a single output with `-o` (#3450, @Cyan4973)
cli: Preserve Permissions and Ownership of regular files (#3432, @felixhandte)
cli: Print zlib/lz4/lzma library versions with `-vv` (#3030, @terrelln)
cli: Print checksum value for single frame files with `-lv` (#3332, @Cyan4973)
cli: Print `dictID` when present with `-lv` (#3184, @htnhan)
cli: when `stderr` is *not* the console, disable status updates, but preserve final summary (#3458, @Cyan4973)
cli: support `--best` and `--no-name` in `gzip` compatibility mode (#3059, @dirkmueller)
cli: support for `posix` high resolution timer `clock_gettime()`, for improved benchmark accuracy (#3423, @Cyan4973)
cli: improved help/usage (`-h`, `-H`) formatting (#3094, @dirkmueller and #3385, @jonpalmisc)
cli: Fix better handling of bogus numeric values (#3268, @ctkhanhly)
cli: Fix input consists of multiple files _and_ `stdin` (#3222, @yoniko)
cli: Fix tiny files passthrough (#3215, @cgbur)
cli: Fix for `-r` on empty directory (#3027, @brailovich)
cli: Fix empty string as argument for `--output-dir-*` (#3220, @embg)
cli: Fix decompression memory usage reported by `-vv --long` (#3042, @u1f35c, and #3232, @zengyijing)
cli: Fix infinite loop when empty input is passed to trainer (#3081, @terrelln)
cli: Fix `--adapt` doesn't work when `--no-progress` is also set (#3354, @terrelln)
api: Support for Block-Level Sequence Producer (#3333, @embg)
api: Support for in-place decompression (#3432, @terrelln)
api: New `ZSTD_CCtx_setCParams()` function, set all parameters defined in a `ZSTD_compressionParameters` structure (#3403, @Cyan4973)
api: Streaming decompression detects incorrect header ID sooner (#3175, @Cyan4973)
api: Window size resizing optimization for edge case (#3345, @daniellerozenblit)
api: More accurate error codes for busy-loop scenarios (#3413, #3455, @Cyan4973)
api: Fix limit overflow in `compressBound` and `decompressBound` (#3362, #3373, Cyan4973) reported by @nigeltao
api: Deprecate several advanced experimental functions: streaming (#3408, @embg), copy (#3196, @mileshu)
bug: Fix corruption that rarely occurs in 32-bit mode with wlog=25 (#3361, @terrelln)
bug: Fix for block-splitter (#3033, @Cyan4973)
bug: Fixes for Sequence Compression API (#3023, #3040, @Cyan4973)
bug: Fix leaking thread handles on Windows (#3147, @animalize)
bug: Fix timing issues with cmake/meson builds (#3166, #3167, #3170, @Cyan4973)
build: Allow user to select legacy level for cmake (#3050, @shadchin)
build: Enable legacy support by default in cmake (#3079, @niamster)
build: Meson build script improvements (#3039, #3120, #3122, #3327, #3357, @eli-schwartz and #3276, @neheb)
build: Add aarch64 to supported architectures for zstd_trace (#3054, @ooosssososos)
build: support AIX architecture (#3219, @qiongsiwu)
build: Fix `ZSTD_LIB_MINIFY` build macro, which now reduces static library size by half (#3366, @terrelln)
build: Fix Windows issues with Multithreading translation layer (#3364, #3380, @yoniko) and ARM64 target (#3320, @cwoffenden)
build: Fix `cmake` script (#3382, #3392, @terrelln and #3252 @Tachi107 and #3167 @Cyan4973)
doc: Updated man page, providing more details for `--train` mode (#3112, @Cyan4973)
doc: Add decompressor errata document (#3092, @terrelln)
misc: Enable Intel CET (#2992, #2994, @hjl-tools)
misc: Fix `contrib/` seekable format (#3058, @yhoogstrate and #3346, @daniellerozenblit)
misc: Improve speed of the one-file library generator (#3241, @wahern and #3005, @cwoffenden)
v1.5.3 (dev version, unpublished)
v1.5.2 (Jan, 2022)
perf: Regain Minimal memset()-ing During Reuse of Compression Contexts (@Cyan4973, #2969)
build: Build Zstd with `noexecstack` on All Architectures (@felixhandte, #2964)
doc: Clarify Licensing (@terrelln, #2981)
v1.5.1 (Dec, 2021)
perf: rebalanced compression levels, to better match the intended speed/level curve, by @senhuang42
perf: faster huffman decoder, using x64 assembly, by @terrelln
perf: slightly faster high speed modes (strategies fast & dfast), by @felixhandte
perf: improved binary size and faster compilation times, by @terrelln
perf: new row64 mode, used notably in level 12, by @senhuang42
perf: faster mid-level compression speed in presence of highly repetitive patterns, by @senhuang42
perf: minor compression ratio improvements for small data at high levels, by @cyan4973
perf: reduced stack usage (mostly useful for Linux Kernel), by @terrelln
perf: faster compression speed on incompressible data, by @bindhvo
perf: on-demand reduced ZSTD_DCtx state size, using build macro ZSTD_DECODER_INTERNAL_BUFFER, at a small cost of performance, by @bindhvo
build: allows hiding static symbols in the dynamic library, using build macro, by @skitt
build: support for m68k (Motorola 68000's), by @cyan4973
build: improved AIX support, by @Helflym
build: improved meson unofficial build, by @eli-schwartz
cli : custom memory limit when training dictionary (#2925), by @embg
cli : report advanced parameters information when compressing in very verbose mode (``-vv`), by @Svetlitski-FB
v1.5.0 (May 11, 2021)
api: Various functions promoted from experimental to stable API: (#2579-2581, @senhuang42)
`ZSTD_defaultCLevel()`
`ZSTD_getDictID_fromCDict()`
api: Several experimental functions have been deprecated and will emit a compiler warning (#2582, @senhuang42)
`ZSTD_compress_advanced()`
`ZSTD_compress_usingCDict_advanced()`
`ZSTD_compressBegin_advanced()`
`ZSTD_compressBegin_usingCDict_advanced()`
`ZSTD_initCStream_srcSize()`
`ZSTD_initCStream_usingDict()`
`ZSTD_initCStream_usingCDict()`
`ZSTD_initCStream_advanced()`
`ZSTD_initCStream_usingCDict_advanced()`
`ZSTD_resetCStream()`
api: ZSTDMT_NBWORKERS_MAX reduced to 64 for 32-bit environments (@Cyan4973)
perf: Significant speed improvements for middle compression levels (#2494, @senhuang42 @terrelln)
perf: Block splitter to improve compression ratio, enabled by default for high compression levels (#2447, @senhuang42)
perf: Decompression loop refactor, speed improvements on `clang` and for `--long` modes (#2614 #2630, @Cyan4973)
perf: Reduced stack usage during compression and decompression entropy stage (#2522 #2524, @terrelln)
bug: Improve setting permissions of created files (#2525, @felixhandte)
bug: Fix large dictionary non-determinism (#2607, @terrelln)
bug: Fix non-determinism test failures on Linux i686 (#2606, @terrelln)
bug: Fix various dedicated dictionary search bugs (#2540 #2586, @senhuang42 @felixhandte)
bug: Ensure `ZSTD_estimateCCtxSize*() `monotonically increases with compression level (#2538, @senhuang42)
bug: Fix --patch-from mode parameter bound bug with small files (#2637, @occivink)
bug: Fix UBSAN error in decompression (#2625, @terrelln)
bug: Fix superblock compression divide by zero bug (#2592, @senhuang42)
bug: Make the number of physical CPU cores detection more robust (#2517, @PaulBone)
doc: Improve `zdict.h` dictionary training API documentation (#2622, @terrelln)
doc: Note that public `ZSTD_free*()` functions accept NULL pointers (#2521, @animalize)
doc: Add style guide docs for open source contributors (#2626, @Cyan4973)
tests: Better regression test coverage for different dictionary modes (#2559, @senhuang42)
tests: Better test coverage of index reduction (#2603, @terrelln)
tests: OSS-Fuzz coverage for seekable format (#2617, @senhuang42)
tests: Test coverage for ZSTD threadpool API (#2604, @senhuang42)
build: Dynamic library built multithreaded by default (#2584, @senhuang42)
build: Move `zstd_errors.h` and `zdict.h` to `lib/` root (#2597, @terrelln)
build: Allow `ZSTDMT_JOBSIZE_MIN` to be configured at compile-time, reduce default to 512KB (#2611, @Cyan4973)
build: Single file library build script moved to `build/` directory (#2618, @felixhandte)
build: `ZBUFF_*()` is no longer built by default (#2583, @senhuang42)
build: Fixed Meson build (#2548, @SupervisedThinking @kloczek)
build: Fix excessive compiler warnings with clang-cl and CMake (#2600, @nickhutchinson)
build: Detect presence of `md5` on Darwin (#2609, @felixhandte)
build: Avoid SIGBUS on armv6 (#2633, @bmwiedmann)
cli: `--progress` flag added to always display progress bar (#2595, @senhuang42)
cli: Allow reading from block devices with `--force` (#2613, @felixhandte)
cli: Fix CLI filesize display bug (#2550, @Cyan4973)
cli: Fix windows CLI `--filelist` end-of-line bug (#2620, @Cyan4973)
contrib: Various fixes for linux kernel patch (#2539, @terrelln)
contrib: Seekable format - Decompression hanging edge case fix (#2516, @senhuang42)
contrib: Seekable format - New seek table-only API (#2113 #2518, @mdittmer @Cyan4973)
contrib: Seekable format - Fix seek table descriptor check when loading (#2534, @foxeng)
contrib: Seekable format - Decompression fix for large offsets, (#2594, @azat)
misc: Automatically published release tarballs available on Github (#2535, @felixhandte)
v1.4.9 (Mar 1, 2021)
bug: Use `umask()` to Constrain Created File Permissions (#2495, @felixhandte)
bug: Make Simple Single-Pass Functions Ignore Advanced Parameters (#2498, @terrelln)
api: Add (De)Compression Tracing Functionality (#2482, @terrelln)
api: Support References to Multiple DDicts (#2446, @senhuang42)
api: Add Function to Generate Skippable Frame (#2439, @senhuang42)
perf: New Algorithms for the Long Distance Matcher (#2483, @mpu)
perf: Performance Improvements for Long Distance Matcher (#2464, @mpu)
perf: Don't Shrink Window Log when Streaming with a Dictionary (#2451, @terrelln)
cli: Fix `--output-dir-mirror`'s Rejection of `..`-Containing Paths (#2512, @felixhandte)
cli: Allow Input From Console When `-f`/`--force` is Passed (#2466, @felixhandte)
cli: Improve Help Message (#2500, @senhuang42)
tests: Remove Flaky Tests (#2455, #2486, #2445, @Cyan4973)
tests: Correctly Invoke md5 Utility on NetBSD (#2492, @niacat)
tests: Avoid Using `stat -c` on NetBSD (#2513, @felixhandte)
build: Zstd CLI Can Now be Linked to Dynamic `libzstd` (#2457, #2454 @Cyan4973)
build: Hide and Avoid Using Static-Only Symbols (#2501, #2504, @skitt)
build: CMake: Enable Only C for lib/ and programs/ Projects (#2498, @concatime)
build: CMake: Use `configure_file()` to Create the `.pc` File (#2462, @lazka)
build: Fix Fuzzer Compiler Detection & Update UBSAN Flags (#2503, @terrelln)
build: Add Guards for `_LARGEFILE_SOURCE` and `_LARGEFILE64_SOURCE` (#2444, @indygreg)
build: Improve `zlibwrapper` Makefile (#2437, @Cyan4973)
contrib: Add `recover_directory` Program (#2473, @terrelln)
doc: Change License Year to 2021 (#2452 & #2465, @terrelln & @senhuang42)
doc: Fix Typos (#2459, @ThomasWaldmann)
v1.4.8 (Dec 18, 2020)
hotfix: wrong alignment of an internal buffer
v1.4.7 (Dec 16, 2020)
perf: stronger --long mode at high compression levels, by @senhuang42
perf: stronger --patch-from at high compression levels, thanks to --long improvements
perf: faster dictionary compression at medium compression levels, by @felixhandte
perf: small speed & memory usage improvements for ZSTD_compress2(), by @terrelln
perf: improved fast compression speeds with Visual Studio, by @animalize
cli : Set nb of threads with environment variable ZSTD_NBTHREADS, by @senhuang42
cli : accept decompressing files with *.zstd suffix
cli : provide a condensed summary by default when processing multiple files
cli : fix : stdin input no longer confused as user prompt
cli : improve accuracy of several error messages
api : new sequence ingestion API, by @senhuang42
api : shared thread pool: control total nb of threads used by multiple compression jobs, by @marxin
api : new ZSTD_getDictID_fromCDict(), by @LuAPi
api : zlibWrapper only uses public API, and is compatible with dynamic library, by @terrelln
api : fix : multithreaded compression has predictable output even in special cases (see #2327) (issue not accessible from cli)
api : fix : dictionary compression correctly respects dictionary compression level (see #2303) (issue not accessible from cli)
build: fix cmake script when using path with spaces, by @terrelln
build: improved compile-time detection of aarch64/neon platforms, by @bsdimp
build: Fix building on AIX 5.1, by @likema
build: compile paramgrill with cmake on Windows, requested by @mirh
doc : clarify repcode updates in format specification, by @felixhandte
v1.4.6
fix : Always return dstSize_tooSmall when that is the case
fix : Fix ZSTD_initCStream_advanced() with static allocation and no dictionary
perf: Improve small block decompression speed by 20%+, by @terrelln
perf: Reduce compression stack usage by 1 KB, by @terrelln
perf: Improve decompression speed by improving ZSTD_wildcopy, by @helloguo (#2252, #2256)
perf: Improve histogram construction, by @cyan4973 (#2253)
cli : Add --output-dir-mirror option, by @xxie24 (#2219)
cli : Warn when (de)compressing multiple files into a single output, by @senhuang42 (#2279)
cli : Improved progress bar and status summary when (de)compressing multiple files, by @senhuang42 (#2283)
cli : Call stat less often, by @felixhandte (#2262)
cli : Allow --patch-from XXX and --filelist XXX in addition to --patch-from=XXX and --filelist=XXX, by @cyan4973 (#2250)
cli : Allow --patch-from to compress stdin with --stream-size, by @bimbashrestha (#2206)
api : Do not install zbuff.h, since it has long been deprecated, by @cyan4973 (#2166).
api : Fix ZSTD_CCtx_setParameter() with ZSTD_c_compressionLevel to make 0 mean default level, by @i-do-cpp (#2291)
api : Rename ZSTDMT_NBTHREADS_MAX to ZSTDMT_NBWORKERS_MAX, by @marxin (#2228).
build: Install pkg-config file with CMake and MinGW, by @tonytheodore (#2183)
build: Install DLL with CMake on Windows, by @BioDataAnalysis (#2221)
build: Fix DLL install location with CMake, by @xantares and @bimbashrestha (#2186)
build: Add ZSTD_NO_UNUSED_FUNCTIONS macro to hide unused functions
build: Add ZSTD_NO_INTRINSICS macro to avoid explicit intrinsics
build: Add STATIC_BMI2 macro for compile time detection of BMI2 on MSVC, by @Niadb (#2258)
build: Fix -Wcomma warnings, by @cwoffenden
build: Remove distutils requirement for meson build, by @neheb (#2197)
build: Fix cli compilation with uclibc
build: Fix cli compilation without st_mtime, by @ffontaine (#2246)
build: Fix shadowing warnings in library
build: Fix single file library compilation with Enscripten, by @yoshihitoh (#2227)
misc: Improve single file library and include dictBuilder, by @cwoffenden
misc: Allow compression dictionaries with missing symbols
misc: Add freestanding translation script in contrib/freestanding_lib
misc: Collect all of zstd's libc dependencies into zstd_deps.h
doc : Add ZSTD_versionString() to manual, by @animalize
doc : Fix documentation for ZSTD_CCtxParams_setParameter(), by @felixhandte (#2270)
v1.4.5 (May 22, 2020)
fix : Compression ratio regression on huge files (> 3 GB) using high levels (--ultra) and multithreading, by @terrelln
perf: Improved decompression speed: x64 : +10% (clang) / +5% (gcc); ARM : from +15% to +50%, depending on SoC, by @terrelln
perf: Automatically downsizes ZSTD_DCtx when too large for too long (#2069, by @bimbashreshta)
perf: Improved fast compression speed on aarch64 (#2040, ~+3%, by @caoyzh)
perf: Small level 1 compression speed gains (depending on compiler)
cli : New --patch-from command, create and apply patches from files, by @bimbashreshta
cli : New --filelist= : Provide a list of files to operate upon from a file
cli : -b -d command can now benchmark decompression on multiple files
cli : New --no-content-size command
cli : New --show-default-cparams information command
api : ZDICT_finalizeDictionary() is promoted to stable (#2111)
api : new experimental parameter ZSTD_d_stableOutBuffer (#2094)
build: Generate a single-file libzstd library (#2065, by @cwoffenden)
build: Relative includes no longer require -I compiler flags for zstd lib subdirs (#2103, by @felixhandte)
build: zstd now compiles cleanly under -pedantic (#2099)
build: zstd now compiles with make-4.3
build: Support mingw cross-compilation from Linux, by @Ericson2314
build: Meson multi-thread build fix on windows
build: Some misc icc fixes backed by new ci test on travis
misc: bitflip analyzer tool, by @felixhandte
misc: Extend largeNbDicts benchmark to compression
misc: Edit-distance match finder in contrib/
doc : Improved beginner CONTRIBUTING.md docs
doc : New issue templates for zstd
v1.4.4 (Nov 6, 2019)
perf: Improved decompression speed, by > 10%, by @terrelln
perf: Better compression speed when re-using a context, by @felixhandte
perf: Fix compression ratio when compressing large files with small dictionary, by @senhuang42
perf: zstd reference encoder can generate RLE blocks, by @bimbashrestha
perf: minor generic speed optimization, by @davidbolvansky
api: new ability to extract sequences from the parser for analysis, by @bimbashrestha
api: fixed decoding of magic-less frames, by @terrelln
api: fixed ZSTD_initCStream_advanced() performance with fast modes, reported by @QrczakMK
cli: Named pipes support, by @bimbashrestha
cli: short tar's extension support, by @stokito
cli: command --output-dir-flat= , generates target files into requested directory, by @senhuang42
cli: commands --stream-size=# and --size-hint=#, by @nmagerko
cli: command --exclude-compressed, by @shashank0791
cli: faster `-t` test mode
cli: improved some error messages, by @vangyzen
cli: fix command `-D dictionary` on Windows, reported by @artyompetrov
cli: fix rare deadlock condition within dictionary builder, by @terrelln
build: single-file decoder with emscripten compilation script, by @cwoffenden
build: fixed zlibWrapper compilation on Visual Studio, reported by @bluenlive
build: fixed deprecation warning for certain gcc version, reported by @jasonma163
build: fix compilation on old gcc versions, by @cemeyer
build: improved installation directories for cmake script, by Dmitri Shubin
pack: modified pkgconfig, for better integration into openwrt, requested by @neheb
misc: Improved documentation : ZSTD_CLEVEL, DYNAMIC_BMI2, ZSTD_CDict, function deprecation, zstd format
misc: fixed educational decoder : accept larger literals section, and removed UNALIGNED() macro
v1.4.3 (Aug 20, 2019)
bug: Fix Dictionary Compression Ratio Regression by @cyan4973 (#1709)
bug: Fix Buffer Overflow in legacy v0.3 decompression by @felixhandte (#1722)
build: Add support for IAR C/C++ Compiler for Arm by @joseph0918 (#1705)
v1.4.2 (Jul 26, 2019)
bug: Fix bug in zstd-0.5 decoder by @terrelln (#1696)
bug: Fix seekable decompression in-memory API by @iburinoc (#1695)
misc: Validate blocks are smaller than size limit by @vivekmg (#1685)
misc: Restructure source files by @ephiepark (#1679)
v1.4.1 (Jul 20, 2019)
bug: Fix data corruption in niche use cases by @terrelln (#1659)
bug: Fuzz legacy modes, fix uncovered bugs by @terrelln (#1593, #1594, #1595)
bug: Fix out of bounds read by @terrelln (#1590)
perf: Improve decode speed by ~7% @mgrice (#1668)
perf: Slightly improved compression ratio of level 3 and 4 (ZSTD_dfast) by @cyan4973 (#1681)
perf: Slightly faster compression speed when re-using a context by @cyan4973 (#1658)
perf: Improve compression ratio for small windowLog by @cyan4973 (#1624)
perf: Faster compression speed in high compression mode for repetitive data by @terrelln (#1635)
api: Add parameter to generate smaller dictionaries by @tyler-tran (#1656)
cli: Recognize symlinks when built in C99 mode by @felixhandte (#1640)
cli: Expose cpu load indicator for each file on -vv mode by @ephiepark (#1631)
cli: Restrict read permissions on destination files by @chungy (#1644)
cli: zstdgrep: handle -f flag by @felixhandte (#1618)
cli: zstdcat: follow symlinks by @vejnar (#1604)
doc: Remove extra size limit on compressed blocks by @felixhandte (#1689)
doc: Fix typo by @yk-tanigawa (#1633)
doc: Improve documentation on streaming buffer sizes by @cyan4973 (#1629)
build: CMake: support building with LZ4 @leeyoung624 (#1626)
build: CMake: install zstdless and zstdgrep by @leeyoung624 (#1647)
build: CMake: respect existing uninstall target by @j301scott (#1619)
build: Make: skip multithread tests when built without support by @michaelforney (#1620)
build: Make: Fix examples/ test target by @sjnam (#1603)
build: Meson: rename options out of deprecated namespace by @lzutao (#1665)
build: Meson: fix build by @lzutao (#1602)
build: Visual Studio: don't export symbols in static lib by @scharan (#1650)
build: Visual Studio: fix linking by @absotively (#1639)
build: Fix MinGW-W64 build by @myzhang1029 (#1600)
misc: Expand decodecorpus coverage by @ephiepark (#1664)
v1.4.0 (Apr 17, 2019)
perf: Improve level 1 compression speed in most scenarios by 6% by @gbtucker and @terrelln
api: Move the advanced API, including all functions in the staging section, to the stable section
api: Make ZSTD_e_flush and ZSTD_e_end block for maximum forward progress
api: Rename ZSTD_CCtxParam_getParameter to ZSTD_CCtxParams_getParameter
api: Rename ZSTD_CCtxParam_setParameter to ZSTD_CCtxParams_setParameter
api: Don't export ZSTDMT functions from the shared library by default
api: Require ZSTD_MULTITHREAD to be defined to use ZSTDMT
api: Add ZSTD_decompressBound() to provide an upper bound on decompressed size by @shakeelrao
api: Fix ZSTD_decompressDCtx() corner cases with a dictionary
api: Move ZSTD_getDictID_*() functions to the stable section
api: Add ZSTD_c_literalCompressionMode flag to enable or disable literal compression by @terrelln
api: Allow compression parameters to be set when a dictionary is used
api: Allow setting parameters before or after ZSTD_CCtx_loadDictionary() is called
api: Fix ZSTD_estimateCStreamSize_usingCCtxParams()
api: Setting ZSTD_d_maxWindowLog to 0 means use the default
cli: Ensure that a dictionary is not used to compress itself by @shakeelrao
cli: Add --[no-]compress-literals flag to enable or disable literal compression
doc: Update the examples to use the advanced API
doc: Explain how to transition from old streaming functions to the advanced API in the header
build: Improve the Windows release packages
build: Improve CMake build by @hjmjohnson
build: Build fixes for FreeBSD by @lwhsu
build: Remove redundant warnings by @thatsafunnyname
build: Fix tests on OpenBSD by @bket
build: Extend fuzzer build system to work with the new clang engine
build: CMake now creates the libzstd.so.1 symlink
build: Improve Menson build by @lzutao
misc: Fix symbolic link detection on FreeBSD
misc: Use physical core count for -T0 on FreeBSD by @cemeyer
misc: Fix zstd --list on truncated files by @kostmo
misc: Improve logging in debug mode by @felixhandte
misc: Add CirrusCI tests by @lwhsu
misc: Optimize dictionary memory usage in corner cases
misc: Improve the dictionary builder on small or homogeneous data
misc: Fix spelling across the repo by @jsoref
v1.3.8 (Dec 28, 2018)
perf: better decompression speed on large files (+7%) and cold dictionaries (+15%)
perf: slightly better compression ratio at high compression modes
api : finalized advanced API, last stage before "stable" status
api : new --rsyncable mode, by @terrelln
api : support decompression of empty frames into NULL (used to be an error) (#1385)
build: new set of macros to build a minimal size decoder, by @felixhandte
build: fix compilation on MIPS32, reported by @clbr (#1441)
build: fix compilation with multiple -arch flags, by @ryandesign
build: highly upgraded meson build, by @lzutao
build: improved buck support, by @obelisk
build: fix cmake script : can create debug build, by @pitrou
build: Makefile : grep works on both colored consoles and systems without color support
build: fixed zstd-pgo, by @bmwiedemann
cli : support ZSTD_CLEVEL environment variable, by @yijinfb (#1423)
cli : --no-progress flag, preserving final summary (#1371), by @terrelln
cli : ensure destination file is not source file (#1422)
cli : clearer error messages, especially when input file not present
doc : clarified zstd_compression_format.md, by @ulikunitz
misc: fixed zstdgrep, returns 1 on failure, by @lzutao
misc: NEWS renamed as CHANGELOG, in accordance with fboss
v1.3.7 (Oct 20, 2018)
perf: slightly better decompression speed on clang (depending on hardware target)
fix : performance of dictionary compression for small input < 4 KB at levels 9 and 10
build: no longer build backtrace by default in release mode; restrict further automatic mode
build: control backtrace support through build macro BACKTRACE
misc: added man pages for zstdless and zstdgrep, by @samrussell
v1.3.6 (Oct 6, 2018)
perf: much faster dictionary builder, by @jenniferliu
perf: faster dictionary compression on small data when using multiple contexts, by @felixhandte
perf: faster dictionary decompression when using a very large number of dictionaries simultaneously
cli : fix : does no longer overwrite destination when source does not exist (#1082)
cli : new command --adapt, for automatic compression level adaptation
api : fix : block api can be streamed with > 4 GB, reported by @catid
api : reduced ZSTD_DDict size by 2 KB
api : minimum negative compression level is defined, and can be queried using ZSTD_minCLevel().
build: support Haiku target, by @korli
build: Read Legacy format is limited to v0.5+ by default. Can be changed at compile time with macro ZSTD_LEGACY_SUPPORT.
doc : zstd_compression_format.md updated to match wording in IETF RFC 8478
misc: tests/paramgrill, a parameter optimizer, by @GeorgeLu97
v1.3.5 (Jun 29, 2018)
perf: much faster dictionary compression, by @felixhandte
perf: small quality improvement for dictionary generation, by @terrelln
perf: slightly improved high compression levels (notably level 19)
mem : automatic memory release for long duration contexts
cli : fix : overlapLog can be manually set
cli : fix : decoding invalid lz4 frames
api : fix : performance degradation for dictionary compression when using advanced API, by @terrelln
api : change : clarify ZSTD_CCtx_reset() vs ZSTD_CCtx_resetParameters(), by @terrelln
build: select custom libzstd scope through control macros, by @GeorgeLu97
build: OpenBSD patch, by @bket
build: make and make all are compatible with -j
doc : clarify zstd_compression_format.md, updated for IETF RFC process
misc: pzstd compatible with reproducible compilation, by @lamby
v1.3.4 (Mar 27, 2018)
perf: faster speed (especially decoding speed) on recent cpus (haswell+)
perf: much better performance associating --long with multi-threading, by @terrelln
perf: better compression at levels 13-15
cli : asynchronous compression by default, for faster experience (use --single-thread for former behavior)
cli : smoother status report in multi-threading mode
cli : added command --fast=#, for faster compression modes
cli : fix crash when not overwriting existing files, by Pádraig Brady (@pixelb)
api : `nbThreads` becomes `nbWorkers` : 1 triggers asynchronous mode
api : compression levels can be negative, for even more speed
api : ZSTD_getFrameProgression() : get precise progress status of ZSTDMT anytime
api : ZSTDMT can accept new compression parameters during compression
api : implemented all advanced dictionary decompression prototypes
build: improved meson recipe, by Shawn Landden (@shawnl)
build: VS2017 scripts, by @HaydnTrigg
misc: all /contrib projects fixed
misc: added /contrib/docker script by @gyscos
v1.3.3 (Dec 21, 2017)
perf: faster zstd_opt strategy (levels 16-19)
fix : bug #944 : multithreading with shared ditionary and large data, reported by @gsliepen
cli : fix : content size written in header by default
cli : fix : improved LZ4 format support, by @felixhandte
cli : new : hidden command `-S`, to benchmark multiple files while generating one result per file
api : fix : support large skippable frames, by @terrelln
api : fix : streaming interface was adding a useless 3-bytes null block to small frames
api : change : when setting `pledgedSrcSize`, use `ZSTD_CONTENTSIZE_UNKNOWN` macro value to mean "unknown"
build: fix : compilation under rhel6 and centos6, reported by @pixelb
build: added `check` target
v1.3.2 (Oct 10, 2017)
new : long range mode, using --long command, by Stella Lau (@stellamplau)
new : ability to generate and decode magicless frames (#591)
changed : maximum nb of threads reduced to 200, to avoid address space exhaustion in 32-bits mode
fix : multi-threading compression works with custom allocators
fix : ZSTD_sizeof_CStream() was over-evaluating memory usage
fix : a rare compression bug when compression generates very large distances and bunch of other conditions (only possible at --ultra -22)
fix : 32-bits build can now decode large offsets (levels 21+)
cli : added LZ4 frame support by default, by Felix Handte (@felixhandte)
cli : improved --list output
cli : new : can split input file for dictionary training, using command -B#
cli : new : clean operation artefact on Ctrl-C interruption
cli : fix : do not change /dev/null permissions when using command -t with root access, reported by @mike155 (#851)
cli : fix : write file size in header in multiple-files mode
api : added macro ZSTD_COMPRESSBOUND() for static allocation
api : experimental : new advanced decompression API
api : fix : sizeof_CCtx() used to over-estimate
build: fix : no-multithread variant compiles without pool.c dependency, reported by Mitchell Blank Jr (@mitchblank) (#819)
build: better compatibility with reproducible builds, by Bernhard M. Wiedemann (@bmwiedemann) (#818)
example : added streaming_memory_usage
license : changed /examples license to BSD + GPLv2
license : fix a few header files to reflect new license (#825)
v1.3.1 (Aug 21, 2017)
New license : BSD + GPLv2
perf: substantially decreased memory usage in Multi-threading mode, thanks to reports by Tino Reichardt (@mcmilk)
perf: Multi-threading supports up to 256 threads. Cap at 256 when more are requested (#760)
cli : improved and fixed --list command, by @ib (#772)
cli : command -vV to list supported formats, by @ib (#771)
build : fixed binary variants, reported by @svenha (#788)
build : fix Visual compilation for non x86/x64 targets, reported by Greg Slazinski (@GregSlazinski) (#718)
API exp : breaking change : ZSTD_getframeHeader() provides more information
API exp : breaking change : pinned down values of error codes
doc : fixed huffman example, by Ulrich Kunitz (@ulikunitz)
new : contrib/adaptive-compression, I/O driven compression strength, by Paul Cruz (@paulcruz74)
new : contrib/long_distance_matching, statistics by Stella Lau (@stellamplau)
updated : contrib/linux-kernel, by Nick Terrell (@terrelln)
v1.3.0 (Jul 6, 2017)
cli : new : `--list` command, by Paul Cruz
cli : changed : xz/lzma support enabled by default
cli : changed : `-t *` continue processing list after a decompression error
API : added : ZSTD_versionString()
API : promoted to stable status : ZSTD_getFrameContentSize(), by Sean Purcell
API exp : new advanced API : ZSTD_compress_generic(), ZSTD_CCtx_setParameter()
API exp : new : API for static or external allocation : ZSTD_initStatic?Ctx()
API exp : added : ZSTD_decompressBegin_usingDDict(), requested by Guy Riddle (#700)
API exp : clarified memory estimation / measurement functions.
API exp : changed : strongest strategy renamed ZSTD_btultra, fastest strategy ZSTD_fast set to 1
tools : decodecorpus can generate random dictionary-compressed samples, by Paul Cruz
new : contrib/seekable_format, demo and API, by Sean Purcell
changed : contrib/linux-kernel, updated version and license, by Nick Terrell
v1.2.0 (May 5, 2017)
cli : changed : Multithreading enabled by default (use target zstd-nomt or HAVE_THREAD=0 to disable)
cli : new : command -T0 means "detect and use nb of cores", by Sean Purcell
cli : new : zstdmt symlink hardwired to `zstd -T0`
cli : new : command --threads=# (#671)
cli : changed : cover dictionary builder by default, for improved quality, by Nick Terrell
cli : new : commands --train-cover and --train-legacy, to select dictionary algorithm and parameters
cli : experimental targets `zstd4` and `xzstd4`, with support for lz4 format, by Sean Purcell
cli : fix : does not output compressed data on console
cli : fix : ignore symbolic links unless --force specified,
API : breaking change : ZSTD_createCDict_advanced(), only use compressionParameters as argument
API : added : prototypes ZSTD_*_usingCDict_advanced(), for direct control over frameParameters.
API : improved: ZSTDMT_compressCCtx() reduced memory usage
API : fix : ZSTDMT_compressCCtx() now provides srcSize in header (#634)
API : fix : src size stored in frame header is controlled at end of frame
API : fix : enforced consistent rules for pledgedSrcSize==0 (#641)
API : fix : error code "GENERIC" replaced by "dstSizeTooSmall" when appropriate
build: improved cmake script, by @Majlen
build: enabled Multi-threading support for *BSD, by Baptiste Daroussin
tools: updated Paramgrill. Command -O# provides best parameters for sample and speed target.
new : contrib/linux-kernel version, by Nick Terrell
v1.1.4 (Mar 18, 2017)
cli : new : can compress in *.gz format, using --format=gzip command, by Przemyslaw Skibinski
cli : new : advanced benchmark command --priority=rt
cli : fix : write on sparse-enabled file systems in 32-bits mode, by @ds77
cli : fix : --rm remains silent when input is stdin
cli : experimental : xzstd, with support for xz/lzma decoding, by Przemyslaw Skibinski
speed : improved decompression speed in streaming mode for single shot scenarios (+5%)
memory: DDict (decompression dictionary) memory usage down from 150 KB to 20 KB
arch: 32-bits variant able to generate and decode very long matches (>32 MB), by Sean Purcell
API : new : ZSTD_findFrameCompressedSize(), ZSTD_getFrameContentSize(), ZSTD_findDecompressedSize()
API : changed : dropped support of legacy versions <= v0.3 (can be changed by modifying ZSTD_LEGACY_SUPPORT value)
build : new: meson build system in contrib/meson, by Dima Krasner
build : improved cmake script, by @Majlen
build : added -Wformat-security flag, as recommended by Padraig Brady
doc : new : educational decoder, by Sean Purcell
v1.1.3 (Feb 7, 2017)
cli : zstd can decompress .gz files (can be disabled with `make zstd-nogz` or `make HAVE_ZLIB=0`)
cli : new : experimental target `make zstdmt`, with multi-threading support
cli : new : improved dictionary builder "cover" (experimental), by Nick Terrell, based on prior work by Giuseppe Ottaviano.
cli : new : advanced commands for detailed parameters, by Przemyslaw Skibinski
cli : fix zstdless on Mac OS-X, by Andrew Janke
cli : fix #232 "compress non-files"
dictBuilder : improved dictionary generation quality, thanks to Nick Terrell
API : new : lib/compress/ZSTDMT_compress.h multithreading API (experimental)
API : new : ZSTD_create?Dict_byReference(), requested by Bartosz Taudul
API : new : ZDICT_finalizeDictionary()
API : fix : ZSTD_initCStream_usingCDict() properly writes dictID into frame header, by Gregory Szorc (#511)
API : fix : all symbols properly exposed in libzstd, by Nick Terrell
build : support for Solaris target, by Przemyslaw Skibinski
doc : clarified specification, by Sean Purcell
v1.1.2 (Dec 15, 2016)
API : streaming : decompression : changed : automatic implicit reset when chain-decoding new frames without init
API : experimental : added : dictID retrieval functions, and ZSTD_initCStream_srcSize()
API : zbuff : changed : prototypes now generate deprecation warnings
lib : improved : faster decompression speed at ultra compression settings and 32-bits mode
lib : changed : only public ZSTD_ symbols are now exposed
lib : changed : reduced usage of stack memory
lib : fixed : several corner case bugs, by Nick Terrell
cli : new : gzstd, experimental version able to decode .gz files, by Przemyslaw Skibinski
cli : new : preserve file attributes
cli : new : added zstdless and zstdgrep tools
cli : fixed : status displays total amount decoded, even for file consisting of multiple frames (like pzstd)
cli : fixed : zstdcat
zlib_wrapper : added support for gz* functions, by Przemyslaw Skibinski
install : better compatibility with FreeBSD, by Dimitry Andric
source tree : changed : zbuff source files moved to lib/deprecated
v1.1.1 (Nov 2, 2016)
New : command -M#, --memory=, --memlimit=, --memlimit-decompress= to limit allowed memory consumption
New : doc/zstd_manual.html, by Przemyslaw Skibinski
Improved : slightly better compression ratio at --ultra levels (>= 20)
Improved : better memory usage when using streaming compression API, thanks to @Rogier-5 report
Added : API : ZSTD_initCStream_usingCDict(), ZSTD_initDStream_usingDDict() (experimental section)
Added : example/multiple_streaming_compression.c
Changed : zstd_errors.h is now installed within /include (and replaces errors_public.h)
Updated man page
Fixed : zstd-small, zstd-compress and zstd-decompress compilation targets
v1.1.0 (Sep 28, 2016)
New : contrib/pzstd, parallel version of zstd, by Nick Terrell
added : NetBSD install target (#338)
Improved : speed for batches of small files
Improved : speed of zlib wrapper, by Przemyslaw Skibinski
Changed : libzstd on Windows supports legacy formats, by Christophe Chevalier
Fixed : CLI -d output to stdout by default when input is stdin (#322)
Fixed : CLI correctly detects console on Mac OS-X
Fixed : CLI supports recursive mode `-r` on Mac OS-X
Fixed : Legacy decoders use unified error codes, reported by benrg (#341), fixed by Przemyslaw Skibinski
Fixed : compatibility with OpenBSD, reported by Juan Francisco Cantero Hurtado (#319)
Fixed : compatibility with Hurd, by Przemyslaw Skibinski (#365)
Fixed : zstd-pgo, reported by octoploid (#329)
v1.0.0 (Sep 1, 2016)
Change Licensing, all project is now BSD, Copyright Facebook
Small decompression speed improvement
API : Streaming API supports legacy format
API : ZDICT_getDictID(), ZSTD_sizeof_{CCtx, DCtx, CStream, DStream}(), ZSTD_setDStreamParameter()
CLI supports legacy formats v0.4+
Fixed : compression fails on certain huge files, reported by Jesse McGrew
Enhanced documentation, by Przemyslaw Skibinski
v0.8.1 (Aug 18, 2016)
New streaming API
Changed : --ultra now enables levels beyond 19
Changed : -i# now selects benchmark time in second
Fixed : ZSTD_compress* can now compress > 4 GB in a single pass, reported by Nick Terrell
Fixed : speed regression on specific patterns (#272)
Fixed : support for Z_SYNC_FLUSH, by Dmitry Krot (#291)
Fixed : ICC compilation, by Przemyslaw Skibinski
v0.8.0 (Aug 2, 2016)
Improved : better speed on clang and gcc -O2, thanks to Eric Biggers
New : Build on FreeBSD and DragonFly, thanks to JrMarino
Changed : modified API : ZSTD_compressEnd()
Fixed : legacy mode with ZSTD_HEAPMODE=0, by Christopher Bergqvist
Fixed : premature end of frame when zero-sized raw block, reported by Eric Biggers
Fixed : large dictionaries (> 384 KB), reported by Ilona Papava
Fixed : checksum correctly checked in single-pass mode
Fixed : combined --test amd --rm, reported by Andreas M. Nilsson
Modified : minor compression level adaptations
Updated : compression format specification to v0.2.0
changed : zstd.h moved to /lib directory
v0.7.5 (Aug 1, 2016)
Transition version, supporting decoding of v0.8.x
v0.7.4 (Jul 17, 2016)
Added : homebrew for Mac, by Daniel Cade
Added : more examples
Fixed : segfault when using small dictionaries, reported by Felix Handte
Modified : default compression level for CLI is now 3
Updated : specification, to v0.1.1
v0.7.3 (Jul 9, 2016)
New : compression format specification
New : `--` separator, stating that all following arguments are file names. Suggested by Chip Turner.
New : `ZSTD_getDecompressedSize()`
New : OpenBSD target, by Juan Francisco Cantero Hurtado
New : `examples` directory
fixed : dictBuilder using HC levels, reported by Bartosz Taudul
fixed : legacy support from ZSTD_decompress_usingDDict(), reported by Felix Handte
fixed : multi-blocks decoding with intermediate uncompressed blocks, reported by Greg Slazinski
modified : removed "mem.h" and "error_public.h" dependencies from "zstd.h" (experimental section)
modified : legacy functions no longer need magic number
v0.7.2 (Jul 4, 2016)
fixed : ZSTD_decompressBlock() using multiple consecutive blocks. Reported by Greg Slazinski.
fixed : potential segfault on very large files (many gigabytes). Reported by Chip Turner.
fixed : CLI displays system error message when destination file cannot be created (#231). Reported by Chip Turner.
v0.7.1 (Jun 23, 2016)
fixed : ZBUFF_compressEnd() called multiple times with too small `dst` buffer, reported by Christophe Chevalier
fixed : dictBuilder fails if first sample is too small, reported by Руслан Ковалёв
fixed : corruption issue, reported by cj
modified : checksum enabled by default in command line mode
v0.7.0 (Jun 17, 2016)
New : Support for directory compression, using `-r`, thanks to Przemyslaw Skibinski
New : Command `--rm`, to remove source file after successful de/compression
New : Visual build scripts, by Christophe Chevalier
New : Support for Sparse File-systems (do not use space for zero-filled sectors)
New : Frame checksum support
New : Support pass-through mode (when using `-df`)
API : more efficient Dictionary API : `ZSTD_compress_usingCDict()`, `ZSTD_decompress_usingDDict()`
API : create dictionary files from custom content, by Giuseppe Ottaviano
API : support for custom malloc/free functions
New : controllable Dictionary ID
New : Support for skippable frames
v0.6.1 (May 13, 2016)
New : zlib wrapper API, thanks to Przemyslaw Skibinski
New : Ability to compile compressor / decompressor separately
Changed : new lib directory structure
Fixed : Legacy codec v0.5 compatible with dictionary decompression
Fixed : Decoder corruption error (#173)
Fixed : null-string roundtrip (#176)
New : benchmark mode can select directory as input
Experimental : midipix support, VMS support
v0.6.0 (Apr 13, 2016)
Stronger high compression modes, thanks to Przemyslaw Skibinski
API : ZSTD_getFrameParams() provides size of decompressed content
New : highest compression modes require `--ultra` command to fully unleash their capacity
Fixed : zstd cli return error code > 0 and removes dst file artifact when decompression fails, thanks to Chip Turner
v0.5.1 (Feb 18, 2016)
New : Optimal parsing => Very high compression modes, thanks to Przemyslaw Skibinski
Changed : Dictionary builder integrated into libzstd and zstd cli
Changed (!) : zstd cli now uses "multiple input files" as default mode. See `zstd -h`.
Fix : high compression modes for big-endian platforms
New : zstd cli : `-t` | `--test` command
v0.5.0 (Feb 5, 2016)
New : dictionary builder utility
Changed : streaming & dictionary API
Improved : better compression of small data
v0.4.7 (Jan 22, 2016)
Improved : small compression speed improvement in HC mode
Changed : `zstd_decompress.c` has ZSTD_LEGACY_SUPPORT to 0 by default
fix : bt search bug
v0.4.6 (Jan 13, 2016)
fix : fast compression mode on Windows
New : cmake configuration file, thanks to Artyom Dymchenko
Improved : high compression mode on repetitive data
New : block-level API
New : ZSTD_duplicateCCtx()
v0.4.5 (Dec 18, 2015)
new : -m/--multiple : compress/decompress multiple files
v0.4.4 (Dec 14, 2015)
Fixed : high compression modes for Windows 32 bits
new : external dictionary API extended to buffered mode and accessible through command line
new : windows DLL project, thanks to Christophe Chevalier
v0.4.3 (Dec 7, 2015)
new : external dictionary API
new : zstd-frugal
v0.4.2 (Dec 2, 2015)
Generic minor improvements for small blocks
Fixed : big-endian compatibility, by Peter Harris (#85)
v0.4.1 (Dec 1, 2015)
Fixed : ZSTD_LEGACY_SUPPORT=0 build mode (reported by Luben)
removed `zstd.c`
v0.4.0 (Nov 29, 2015)
Command line utility compatible with high compression levels
Removed zstdhc => merged into zstd
Added : ZBUFF API (see zstd_buffered.h)
Rolling buffer support
v0.3.6 (Nov 10, 2015)
small blocks params
v0.3.5 (Nov 9, 2015)
minor generic compression improvements
v0.3.4 (Nov 6, 2015)
Faster fast cLevels
v0.3.3 (Nov 5, 2015)
Small compression ratio improvement
v0.3.2 (Nov 2, 2015)
Fixed Visual Studio
v0.3.1 (Nov 2, 2015)
Small compression ratio improvement
v0.3 (Oct 30, 2015)
HC mode : compression levels 2-26
v0.2.2 (Oct 28, 2015)
Fix : Visual Studio 2013 & 2015 release compilation, by Christophe Chevalier
v0.2.1 (Oct 24, 2015)
Fix : Read errors, advanced fuzzer tests, by Hanno Böck
v0.2.0 (Oct 22, 2015)
**Breaking format change**
Faster decompression speed
Can still decode v0.1 format
v0.1.3 (Oct 15, 2015)
fix uninitialization warning, reported by Evan Nemerson
v0.1.2 (Sep 11, 2015)
frame concatenation support
v0.1.1 (Aug 27, 2015)
fix compression bug
detects write-flush errors
v0.1.0 (Aug 25, 2015)
first release

5
raytracer/nvpro_core/third_party/zstd/CODE_OF_CONDUCT.md vendored Normal file
View file

@ -0,0 +1,5 @@
# Code of Conduct
Facebook has adopted a Code of Conduct that we expect project participants to adhere to.
Please read the [full text](https://code.fb.com/codeofconduct/)
so that you can understand what actions will and will not be tolerated.

489
raytracer/nvpro_core/third_party/zstd/CONTRIBUTING.md vendored Normal file
View file

@ -0,0 +1,489 @@
# Contributing to Zstandard
We want to make contributing to this project as easy and transparent as
possible.
## Our Development Process
New versions are being developed in the "dev" branch,
or in their own feature branch.
When they are deemed ready for a release, they are merged into "release".
As a consequence, all contributions must stage first through "dev"
or their own feature branch.
## Pull Requests
We actively welcome your pull requests.
1. Fork the repo and create your branch from `dev`.
2. If you've added code that should be tested, add tests.
3. If you've changed APIs, update the documentation.
4. Ensure the test suite passes.
5. Make sure your code lints.
6. If you haven't already, complete the Contributor License Agreement ("CLA").
## Contributor License Agreement ("CLA")
In order to accept your pull request, we need you to submit a CLA. You only need
to do this once to work on any of Facebook's open source projects.
Complete your CLA here: <https://code.facebook.com/cla>
## Workflow
Zstd uses a branch-based workflow for making changes to the codebase. Typically, zstd
will use a new branch per sizable topic. For smaller changes, it is okay to lump multiple
related changes into a branch.
Our contribution process works in three main stages:
1. Local development
* Update:
* Checkout your fork of zstd if you have not already
```
git checkout https://github.com/<username>/zstd
cd zstd
```
* Update your local dev branch
```
git pull https://github.com/facebook/zstd dev
git push origin dev
```
* Topic and development:
* Make a new branch on your fork about the topic you're developing for
```
# branch names should be concise but sufficiently informative
git checkout -b <branch-name>
git push origin <branch-name>
```
* Make commits and push
```
# make some changes =
git add -u && git commit -m <message>
git push origin <branch-name>
```
* Note: run local tests to ensure that your changes didn't break existing functionality
* Quick check
```
make shortest
```
* Longer check
```
make test
```
2. Code Review and CI tests
* Ensure CI tests pass:
* Before sharing anything to the community, create a pull request in your own fork against the dev branch
and make sure that all GitHub Actions CI tests pass. See the Continuous Integration section below for more information.
* Ensure that static analysis passes on your development machine. See the Static Analysis section
below to see how to do this.
* Create a pull request:
* When you are ready to share you changes to the community, create a pull request from your branch
to facebook:dev. You can do this very easily by clicking 'Create Pull Request' on your fork's home
page.
* From there, select the branch where you made changes as your source branch and facebook:dev
as the destination.
* Examine the diff presented between the two branches to make sure there is nothing unexpected.
* Write a good pull request description:
* While there is no strict template that our contributors follow, we would like them to
sufficiently summarize and motivate the changes they are proposing. We recommend all pull requests,
at least indirectly, address the following points.
* Is this pull request important and why?
* Is it addressing an issue? If so, what issue? (provide links for convenience please)
* Is this a new feature? If so, why is it useful and/or necessary?
* Are there background references and documents that reviewers should be aware of to properly assess this change?
* Note: make sure to point out any design and architectural decisions that you made and the rationale behind them.
* Note: if you have been working with a specific user and would like them to review your work, make sure you mention them using (@<username>)
* Submit the pull request and iterate with feedback.
3. Merge and Release
* Getting approval:
* You will have to iterate on your changes with feedback from other collaborators to reach a point
where your pull request can be safely merged.
* To avoid too many comments on style and convention, make sure that you have a
look at our style section below before creating a pull request.
* Eventually, someone from the zstd team will approve your pull request and not long after merge it into
the dev branch.
* Housekeeping:
* Most PRs are linked with one or more Github issues. If this is the case for your PR, make sure
the corresponding issue is mentioned. If your change 'fixes' or completely addresses the
issue at hand, then please indicate this by requesting that an issue be closed by commenting.
* Just because your changes have been merged does not mean the topic or larger issue is complete. Remember
that the change must make it to an official zstd release for it to be meaningful. We recommend
that contributors track the activity on their pull request and corresponding issue(s) page(s) until
their change makes it to the next release of zstd. Users will often discover bugs in your code or
suggest ways to refine and improve your initial changes even after the pull request is merged.
## Static Analysis
Static analysis is a process for examining the correctness or validity of a program without actually
executing it. It usually helps us find many simple bugs. Zstd uses clang's `scan-build` tool for
static analysis. You can install it by following the instructions for your OS on https://clang-analyzer.llvm.org/scan-build.
Once installed, you can ensure that our static analysis tests pass on your local development machine
by running:
```
make staticAnalyze
```
In general, you can use `scan-build` to static analyze any build script. For example, to static analyze
just `contrib/largeNbDicts` and nothing else, you can run:
```
scan-build make -C contrib/largeNbDicts largeNbDicts
```
### Pitfalls of static analysis
`scan-build` is part of our regular CI suite. Other static analyzers are not.
It can be useful to look at additional static analyzers once in a while (and we do), but it's not a good idea to multiply the nb of analyzers run continuously at each commit and PR. The reasons are :
- Static analyzers are full of false positive. The signal to noise ratio is actually pretty low.
- A good CI policy is "zero-warning tolerance". That means that all issues must be solved, including false positives. This quickly becomes a tedious workload.
- Multiple static analyzers will feature multiple kind of false positives, sometimes applying to the same code but in different ways leading to :
+ tortuous code, trying to please multiple constraints, hurting readability and therefore maintenance. Sometimes, such complexity introduce other more subtle bugs, that are just out of scope of the analyzers.
+ sometimes, these constraints are mutually exclusive : if one try to solve one, the other static analyzer will complain, they can't be both happy at the same time.
- As if that was not enough, the list of false positives change with each version. It's hard enough to follow one static analyzer, but multiple ones with their own update agenda, this quickly becomes a massive velocity reducer.
This is different from running a static analyzer once in a while, looking at the output, and __cherry picking__ a few warnings that seem helpful, either because they detected a genuine risk of bug, or because it helps expressing the code in a way which is more readable or more difficult to misuse. These kinds of reports can be useful, and are accepted.
## Continuous Integration
CI tests run every time a pull request (PR) is created or updated. The exact tests
that get run will depend on the destination branch you specify. Some tests take
longer to run than others. Currently, our CI is set up to run a short
series of tests when creating a PR to the dev branch and a longer series of tests
when creating a PR to the release branch. You can look in the configuration files
of the respective CI platform for more information on what gets run when.
Most people will just want to create a PR with the destination set to their local dev
branch of zstd. You can then find the status of the tests on the PR's page. You can also
re-run tests and cancel running tests from the PR page or from the respective CI's dashboard.
Almost all of zstd's CI runs on GitHub Actions (configured at `.github/workflows`), which will automatically run on PRs to your
own fork. A small number of tests run on other services (e.g. Travis CI, Circle CI, Appveyor).
These require work to set up on your local fork, and (at least for Travis CI) cost money.
Therefore, if the PR on your local fork passes GitHub Actions, feel free to submit a PR
against the main repo.
### Third-party CI
A small number of tests cannot run on GitHub Actions, or have yet to be migrated.
For these, we use a variety of third-party services (listed below). It is not necessary to set
these up on your fork in order to contribute to zstd; however, we do link to instructions for those
who want earlier signal.
| Service | Purpose | Setup Links | Config Path |
|-----------|------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------|
| Travis CI | Used for testing on non-x86 architectures such as PowerPC | https://docs.travis-ci.com/user/tutorial/#to-get-started-with-travis-ci-using-github <br> https://github.com/marketplace/travis-ci | `.travis.yml` |
| AppVeyor | Used for some Windows testing (e.g. cygwin, mingw) | https://www.appveyor.com/blog/2018/10/02/github-apps-integration/ <br> https://github.com/marketplace/appveyor | `appveyor.yml` |
| Cirrus CI | Used for testing on FreeBSD | https://github.com/marketplace/cirrus-ci/ | `.cirrus.yml` |
| Circle CI | Historically was used to provide faster signal,<br/> but we may be able to migrate these to Github Actions | https://circleci.com/docs/2.0/getting-started/#setting-up-circleci <br> https://youtu.be/Js3hMUsSZ2c <br> https://circleci.com/docs/2.0/enable-checks/ | `.circleci/config.yml` |
Note: the instructions linked above mostly cover how to set up a repository with CI from scratch.
The general idea should be the same for setting up CI on your fork of zstd, but you may have to
follow slightly different steps. In particular, please ignore any instructions related to setting up
config files (since zstd already has configs for each of these services).
## Performance
Performance is extremely important for zstd and we only merge pull requests whose performance
landscape and corresponding trade-offs have been adequately analyzed, reproduced, and presented.
This high bar for performance means that every PR which has the potential to
impact performance takes a very long time for us to properly review. That being said, we
always welcome contributions to improve performance (or worsen performance for the trade-off of
something else). Please keep the following in mind before submitting a performance related PR:
1. Zstd isn't as old as gzip but it has been around for time now and its evolution is
very well documented via past Github issues and pull requests. It may be the case that your
particular performance optimization has already been considered in the past. Please take some
time to search through old issues and pull requests using keywords specific to your
would-be PR. Of course, just because a topic has already been discussed (and perhaps rejected
on some grounds) in the past, doesn't mean it isn't worth bringing up again. But even in that case,
it will be helpful for you to have context from that topic's history before contributing.
2. The distinction between noise and actual performance gains can unfortunately be very subtle
especially when microbenchmarking extremely small wins or losses. The only remedy to getting
something subtle merged is extensive benchmarking. You will be doing us a great favor if you
take the time to run extensive, long-duration, and potentially cross-(os, platform, process, etc)
benchmarks on your end before submitting a PR. Of course, you will not be able to benchmark
your changes on every single processor and os out there (and neither will we) but do that best
you can:) We've added some things to think about when benchmarking below in the Benchmarking
Performance section which might be helpful for you.
3. Optimizing performance for a certain OS, processor vendor, compiler, or network system is a perfectly
legitimate thing to do as long as it does not harm the overall performance health of Zstd.
This is a hard balance to strike but please keep in mind other aspects of Zstd when
submitting changes that are clang-specific, windows-specific, etc.
## Benchmarking Performance
Performance microbenchmarking is a tricky subject but also essential for Zstd. We value empirical
testing over theoretical speculation. This guide it not perfect but for most scenarios, it
is a good place to start.
### Stability
Unfortunately, the most important aspect in being able to benchmark reliably is to have a stable
benchmarking machine. A virtual machine, a machine with shared resources, or your laptop
will typically not be stable enough to obtain reliable benchmark results. If you can get your
hands on a desktop, this is usually a better scenario.
Of course, benchmarking can be done on non-hyper-stable machines as well. You will just have to
do a little more work to ensure that you are in fact measuring the changes you've made not and
noise. Here are some things you can do to make your benchmarks more stable:
1. The most simple thing you can do to drastically improve the stability of your benchmark is
to run it multiple times and then aggregate the results of those runs. As a general rule of
thumb, the smaller the change you are trying to measure, the more samples of benchmark runs
you will have to aggregate over to get reliable results. Here are some additional things to keep in
mind when running multiple trials:
* How you aggregate your samples are important. You might be tempted to use the mean of your
results. While this is certainly going to be a more stable number than a raw single sample
benchmark number, you might have more luck by taking the median. The mean is not robust to
outliers whereas the median is. Better still, you could simply take the fastest speed your
benchmark achieved on each run since that is likely the fastest your process will be
capable of running your code. In our experience, this (aggregating by just taking the sample
with the fastest running time) has been the most stable approach.
* The more samples you have, the more stable your benchmarks should be. You can verify
your improved stability by looking at the size of your confidence intervals as you
increase your sample count. These should get smaller and smaller. Eventually hopefully
smaller than the performance win you are expecting.
* Most processors will take some time to get `hot` when running anything. The observations
you collect during that time period will very different from the true performance number. Having
a very large number of sample will help alleviate this problem slightly but you can also
address is directly by simply not including the first `n` iterations of your benchmark in
your aggregations. You can determine `n` by simply looking at the results from each iteration
and then hand picking a good threshold after which the variance in results seems to stabilize.
2. You cannot really get reliable benchmarks if your host machine is simultaneously running
another cpu/memory-intensive application in the background. If you are running benchmarks on your
personal laptop for instance, you should close all applications (including your code editor and
browser) before running your benchmarks. You might also have invisible background applications
running. You can see what these are by looking at either Activity Monitor on Mac or Task Manager
on Windows. You will get more stable benchmark results of you end those processes as well.
* If you have multiple cores, you can even run your benchmark on a reserved core to prevent
pollution from other OS and user processes. There are a number of ways to do this depending
on your OS:
* On linux boxes, you have use https://github.com/lpechacek/cpuset.
* On Windows, you can "Set Processor Affinity" using https://www.thewindowsclub.com/processor-affinity-windows
* On Mac, you can try to use their dedicated affinity API https://developer.apple.com/library/archive/releasenotes/Performance/RN-AffinityAPI/#//apple_ref/doc/uid/TP40006635-CH1-DontLinkElementID_2
3. To benchmark, you will likely end up writing a separate c/c++ program that will link libzstd.
Dynamically linking your library will introduce some added variation (not a large amount but
definitely some). Statically linking libzstd will be more stable. Static libraries should
be enabled by default when building zstd.
4. Use a profiler with a good high resolution timer. See the section below on profiling for
details on this.
5. Disable frequency scaling, turbo boost and address space randomization (this will vary by OS)
6. Try to avoid storage. On some systems you can use tmpfs. Putting the program, inputs and outputs on
tmpfs avoids touching a real storage system, which can have a pretty big variability.
Also check our LLVM's guide on benchmarking here: https://llvm.org/docs/Benchmarking.html
### Zstd benchmark
The fastest signal you can get regarding your performance changes is via the in-build zstd cli
bench option. You can run Zstd as you typically would for your scenario using some set of options
and then additionally also specify the `-b#` option. Doing this will run our benchmarking pipeline
for that options you have just provided. If you want to look at the internals of how this
benchmarking script works, you can check out programs/benchzstd.c
For example: say you have made a change that you believe improves the speed of zstd level 1. The
very first thing you should use to assess whether you actually achieved any sort of improvement
is `zstd -b`. You might try to do something like this. Note: you can use the `-i` option to
specify a running time for your benchmark in seconds (default is 3 seconds).
Usually, the longer the running time, the more stable your results will be.
```
$ git checkout <commit-before-your-change>
$ make && cp zstd zstd-old
$ git checkout <commit-after-your-change>
$ make && cp zstd zstd-new
$ zstd-old -i5 -b1 <your-test-data>
1<your-test-data> : 8990 -> 3992 (2.252), 302.6 MB/s , 626.4 MB/s
$ zstd-new -i5 -b1 <your-test-data>
1<your-test-data> : 8990 -> 3992 (2.252), 302.8 MB/s , 628.4 MB/s
```
Unless your performance win is large enough to be visible despite the intrinsic noise
on your computer, benchzstd alone will likely not be enough to validate the impact of your
changes. For example, the results of the example above indicate that effectively nothing
changed but there could be a small <3% improvement that the noise on the host machine
obscured. So unless you see a large performance win (10-15% consistently) using just
this method of evaluation will not be sufficient.
### Profiling
There are a number of great profilers out there. We're going to briefly mention how you can
profile your code using `instruments` on mac, `perf` on linux and `visual studio profiler`
on Windows.
Say you have an idea for a change that you think will provide some good performance gains
for level 1 compression on Zstd. Typically this means, you have identified a section of
code that you think can be made to run faster.
The first thing you will want to do is make sure that the piece of code is actually taking up
a notable amount of time to run. It is usually not worth optimizing something which accounts for less than
0.0001% of the total running time. Luckily, there are tools to help with this.
Profilers will let you see how much time your code spends inside a particular function.
If your target code snippet is only part of a function, it might be worth trying to
isolate that snippet by moving it to its own function (this is usually not necessary but
might be).
Most profilers (including the profilers discussed below) will generate a call graph of
functions for you. Your goal will be to find your function of interest in this call graph
and then inspect the time spent inside of it. You might also want to look at the annotated
assembly which most profilers will provide you with.
#### Instruments
We will once again consider the scenario where you think you've identified a piece of code
whose performance can be improved upon. Follow these steps to profile your code using
Instruments.
1. Open Instruments
2. Select `Time Profiler` from the list of standard templates
3. Close all other applications except for your instruments window and your terminal
4. Run your benchmarking script from your terminal window
* You will want a benchmark that runs for at least a few seconds (5 seconds will
usually be long enough). This way the profiler will have something to work with
and you will have ample time to attach your profiler to this process:)
* I will just use benchzstd as my benchmarmking script for this example:
```
$ zstd -b1 -i5 <my-data> # this will run for 5 seconds
```
5. Once you run your benchmarking script, switch back over to instruments and attach your
process to the time profiler. You can do this by:
* Clicking on the `All Processes` drop down in the top left of the toolbar.
* Selecting your process from the dropdown. In my case, it is just going to be labeled
`zstd`
* Hitting the bright red record circle button on the top left of the toolbar
6. You profiler will now start collecting metrics from your benchmarking script. Once
you think you have collected enough samples (usually this is the case after 3 seconds of
recording), stop your profiler.
7. Make sure that in toolbar of the bottom window, `profile` is selected.
8. You should be able to see your call graph.
* If you don't see the call graph or an incomplete call graph, make sure you have compiled
zstd and your benchmarking script using debug flags. On mac and linux, this just means
you will have to supply the `-g` flag alone with your build script. You might also
have to provide the `-fno-omit-frame-pointer` flag
9. Dig down the graph to find your function call and then inspect it by double clicking
the list item. You will be able to see the annotated source code and the assembly side by
side.
#### Perf
This wiki has a pretty detailed tutorial on getting started working with perf so we'll
leave you to check that out of you're getting started:
https://perf.wiki.kernel.org/index.php/Tutorial
Some general notes on perf:
* Use `perf stat -r # <bench-program>` to quickly get some relevant timing and
counter statistics. Perf uses a high resolution timer and this is likely one
of the first things your team will run when assessing your PR.
* Perf has a long list of hardware counters that can be viewed with `perf --list`.
When measuring optimizations, something worth trying is to make sure the hardware
counters you expect to be impacted by your change are in fact being so. For example,
if you expect the L1 cache misses to decrease with your change, you can look at the
counter `L1-dcache-load-misses`
* Perf hardware counters will not work on a virtual machine.
#### Visual Studio
TODO
## Issues
We use GitHub issues to track public bugs. Please ensure your description is
clear and has sufficient instructions to be able to reproduce the issue.
Facebook has a [bounty program](https://www.facebook.com/whitehat/) for the safe
disclosure of security bugs. In those cases, please go through the process
outlined on that page and do not file a public issue.
## Coding Style
It's a pretty long topic, which is difficult to summarize in a single paragraph.
As a rule of thumbs, try to imitate the coding style of
similar lines of codes around your contribution.
The following is a non-exhaustive list of rules employed in zstd code base:
### C90
This code base is following strict C90 standard,
with 2 extensions : 64-bit `long long` types, and variadic macros.
This rule is applied strictly to code within `lib/` and `programs/`.
Sub-project in `contrib/` are allowed to use other conventions.
### C++ direct compatibility : symbol mangling
All public symbol declarations must be wrapped in `extern “C” { … }`,
so that this project can be compiled as C++98 code,
and linked into C++ applications.
### Minimal Frugal
This design requirement is fundamental to preserve the portability of the code base.
#### Dependencies
- Reduce dependencies to the minimum possible level.
Any dependency should be considered “bad” by default,
and only tolerated because it provides a service in a better way than can be achieved locally.
The only external dependencies this repository tolerates are
standard C libraries, and in rare cases, system level headers.
- Within `lib/`, this policy is even more drastic.
The only external dependencies allowed are `<assert.h>`, `<stdlib.h>`, `<string.h>`,
and even then, not directly.
In particular, no function shall ever allocate on heap directly,
and must use instead `ZSTD_malloc()` and equivalent.
Other accepted non-symbol headers are `<stddef.h>` and `<limits.h>`.
- Within the project, there is a strict hierarchy of dependencies that must be respected.
`programs/` is allowed to depend on `lib/`, but only its public API.
Within `lib/`, `lib/common` doesn't depend on any other directory.
`lib/compress` and `lib/decompress` shall not depend on each other.
`lib/dictBuilder` can depend on `lib/common` and `lib/compress`, but not `lib/decompress`.
#### Resources
- Functions in `lib/` must use very little stack space,
several dozens of bytes max.
Everything larger must use the heap allocator,
or require a scratch buffer to be emplaced manually.
### Naming
* All public symbols are prefixed with `ZSTD_`
+ private symbols, with a scope limited to their own unit, are free of this restriction.
However, since `libzstd` source code can be amalgamated,
each symbol name must attempt to be (and remain) unique.
Avoid too generic names that could become ground for future collisions.
This generally implies usage of some form of prefix.
* For symbols (functions and variables), naming convention is `PREFIX_camelCase`.
+ In some advanced cases, one can also find :
- `PREFIX_prefix2_camelCase`
- `PREFIX_camelCase_extendedQualifier`
* Multi-words names generally consist of an action followed by object:
- for example : `ZSTD_createCCtx()`
* Prefer positive actions
- `goBackward` rather than `notGoForward`
* Type names (`struct`, etc.) follow similar convention,
except that they are allowed and even invited to start by an Uppercase letter.
Example : `ZSTD_CCtx`, `ZSTD_CDict`
* Macro names are all Capital letters.
The same composition rules (`PREFIX_NAME_QUALIFIER`) apply.
* File names are all lowercase letters.
The convention is `snake_case`.
File names **must** be unique across the entire code base,
even when they stand in clearly separated directories.
### Qualifiers
* This code base is `const` friendly, if not `const` fanatical.
Any variable that can be `const` (aka. read-only) **must** be `const`.
Any pointer which content will not be modified must be `const`.
This property is then controlled at compiler level.
`const` variables are an important signal to readers that this variable isn't modified.
Conversely, non-const variables are a signal to readers to watch out for modifications later on in the function.
* If a function must be inlined, mention it explicitly,
using project's own portable macros, such as `FORCE_INLINE_ATTR`,
defined in `lib/common/compiler.h`.
### Debugging
* **Assertions** are welcome, and should be used very liberally,
to control any condition the code expects for its correct execution.
These assertion checks will be run in debug builds, and disabled in production.
* For traces, this project provides its own debug macros,
in particular `DEBUGLOG(level, ...)`, defined in `lib/common/debug.h`.
### Code documentation
* Avoid code documentation that merely repeats what the code is already stating.
Whenever applicable, prefer employing the code as the primary way to convey explanations.
Example 1 : `int nbTokens = n;` instead of `int i = n; /* i is a nb of tokens *./`.
Example 2 : `assert(size > 0);` instead of `/* here, size should be positive */`.
* At declaration level, the documentation explains how to use the function or variable
and when applicable why it's needed, of the scenarios where it can be useful.
* At implementation level, the documentation explains the general outline of the algorithm employed,
and when applicable why this specific choice was preferred.
### General layout
* 4 spaces for indentation rather than tabs
* Code documentation shall directly precede function declaration or implementation
* Function implementations and its code documentation should be preceded and followed by an empty line
## License
By contributing to Zstandard, you agree that your contributions will be licensed
under both the [LICENSE](LICENSE) file and the [COPYING](COPYING) file in the root directory of this source tree.

339
raytracer/nvpro_core/third_party/zstd/COPYING vendored Normal file
View file

@ -0,0 +1,339 @@
GNU GENERAL PUBLIC LICENSE
Version 2, June 1991
Copyright (C) 1989, 1991 Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
Preamble
The licenses for most software are designed to take away your
freedom to share and change it. By contrast, the GNU General Public
License is intended to guarantee your freedom to share and change free
software--to make sure the software is free for all its users. This
General Public License applies to most of the Free Software
Foundation's software and to any other program whose authors commit to
using it. (Some other Free Software Foundation software is covered by
the GNU Lesser General Public License instead.) You can apply it to
your programs, too.
When we speak of free software, we are referring to freedom, not
price. Our General Public Licenses are designed to make sure that you
have the freedom to distribute copies of free software (and charge for
this service if you wish), that you receive source code or can get it
if you want it, that you can change the software or use pieces of it
in new free programs; and that you know you can do these things.
To protect your rights, we need to make restrictions that forbid
anyone to deny you these rights or to ask you to surrender the rights.
These restrictions translate to certain responsibilities for you if you
distribute copies of the software, or if you modify it.
For example, if you distribute copies of such a program, whether
gratis or for a fee, you must give the recipients all the rights that
you have. You must make sure that they, too, receive or can get the
source code. And you must show them these terms so they know their
rights.
We protect your rights with two steps: (1) copyright the software, and
(2) offer you this license which gives you legal permission to copy,
distribute and/or modify the software.
Also, for each author's protection and ours, we want to make certain
that everyone understands that there is no warranty for this free
software. If the software is modified by someone else and passed on, we
want its recipients to know that what they have is not the original, so
that any problems introduced by others will not reflect on the original
authors' reputations.
Finally, any free program is threatened constantly by software
patents. We wish to avoid the danger that redistributors of a free
program will individually obtain patent licenses, in effect making the
program proprietary. To prevent this, we have made it clear that any
patent must be licensed for everyone's free use or not licensed at all.
The precise terms and conditions for copying, distribution and
modification follow.
GNU GENERAL PUBLIC LICENSE
TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
0. This License applies to any program or other work which contains
a notice placed by the copyright holder saying it may be distributed
under the terms of this General Public License. The "Program", below,
refers to any such program or work, and a "work based on the Program"
means either the Program or any derivative work under copyright law:
that is to say, a work containing the Program or a portion of it,
either verbatim or with modifications and/or translated into another
language. (Hereinafter, translation is included without limitation in
the term "modification".) Each licensee is addressed as "you".
Activities other than copying, distribution and modification are not
covered by this License; they are outside its scope. The act of
running the Program is not restricted, and the output from the Program
is covered only if its contents constitute a work based on the
Program (independent of having been made by running the Program).
Whether that is true depends on what the Program does.
1. You may copy and distribute verbatim copies of the Program's
source code as you receive it, in any medium, provided that you
conspicuously and appropriately publish on each copy an appropriate
copyright notice and disclaimer of warranty; keep intact all the
notices that refer to this License and to the absence of any warranty;
and give any other recipients of the Program a copy of this License
along with the Program.
You may charge a fee for the physical act of transferring a copy, and
you may at your option offer warranty protection in exchange for a fee.
2. You may modify your copy or copies of the Program or any portion
of it, thus forming a work based on the Program, and copy and
distribute such modifications or work under the terms of Section 1
above, provided that you also meet all of these conditions:
a) You must cause the modified files to carry prominent notices
stating that you changed the files and the date of any change.
b) You must cause any work that you distribute or publish, that in
whole or in part contains or is derived from the Program or any
part thereof, to be licensed as a whole at no charge to all third
parties under the terms of this License.
c) If the modified program normally reads commands interactively
when run, you must cause it, when started running for such
interactive use in the most ordinary way, to print or display an
announcement including an appropriate copyright notice and a
notice that there is no warranty (or else, saying that you provide
a warranty) and that users may redistribute the program under
these conditions, and telling the user how to view a copy of this
License. (Exception: if the Program itself is interactive but
does not normally print such an announcement, your work based on
the Program is not required to print an announcement.)
These requirements apply to the modified work as a whole. If
identifiable sections of that work are not derived from the Program,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works. But when you
distribute the same sections as part of a whole which is a work based
on the Program, the distribution of the whole must be on the terms of
this License, whose permissions for other licensees extend to the
entire whole, and thus to each and every part regardless of who wrote it.
Thus, it is not the intent of this section to claim rights or contest
your rights to work written entirely by you; rather, the intent is to
exercise the right to control the distribution of derivative or
collective works based on the Program.
In addition, mere aggregation of another work not based on the Program
with the Program (or with a work based on the Program) on a volume of
a storage or distribution medium does not bring the other work under
the scope of this License.
3. You may copy and distribute the Program (or a work based on it,
under Section 2) in object code or executable form under the terms of
Sections 1 and 2 above provided that you also do one of the following:
a) Accompany it with the complete corresponding machine-readable
source code, which must be distributed under the terms of Sections
1 and 2 above on a medium customarily used for software interchange; or,
b) Accompany it with a written offer, valid for at least three
years, to give any third party, for a charge no more than your
cost of physically performing source distribution, a complete
machine-readable copy of the corresponding source code, to be
distributed under the terms of Sections 1 and 2 above on a medium
customarily used for software interchange; or,
c) Accompany it with the information you received as to the offer
to distribute corresponding source code. (This alternative is
allowed only for noncommercial distribution and only if you
received the program in object code or executable form with such
an offer, in accord with Subsection b above.)
The source code for a work means the preferred form of the work for
making modifications to it. For an executable work, complete source
code means all the source code for all modules it contains, plus any
associated interface definition files, plus the scripts used to
control compilation and installation of the executable. However, as a
special exception, the source code distributed need not include
anything that is normally distributed (in either source or binary
form) with the major components (compiler, kernel, and so on) of the
operating system on which the executable runs, unless that component
itself accompanies the executable.
If distribution of executable or object code is made by offering
access to copy from a designated place, then offering equivalent
access to copy the source code from the same place counts as
distribution of the source code, even though third parties are not
compelled to copy the source along with the object code.
4. You may not copy, modify, sublicense, or distribute the Program
except as expressly provided under this License. Any attempt
otherwise to copy, modify, sublicense or distribute the Program is
void, and will automatically terminate your rights under this License.
However, parties who have received copies, or rights, from you under
this License will not have their licenses terminated so long as such
parties remain in full compliance.
5. You are not required to accept this License, since you have not
signed it. However, nothing else grants you permission to modify or
distribute the Program or its derivative works. These actions are
prohibited by law if you do not accept this License. Therefore, by
modifying or distributing the Program (or any work based on the
Program), you indicate your acceptance of this License to do so, and
all its terms and conditions for copying, distributing or modifying
the Program or works based on it.
6. Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the
original licensor to copy, distribute or modify the Program subject to
these terms and conditions. You may not impose any further
restrictions on the recipients' exercise of the rights granted herein.
You are not responsible for enforcing compliance by third parties to
this License.
7. If, as a consequence of a court judgment or allegation of patent
infringement or for any other reason (not limited to patent issues),
conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License. If you cannot
distribute so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you
may not distribute the Program at all. For example, if a patent
license would not permit royalty-free redistribution of the Program by
all those who receive copies directly or indirectly through you, then
the only way you could satisfy both it and this License would be to
refrain entirely from distribution of the Program.
If any portion of this section is held invalid or unenforceable under
any particular circumstance, the balance of the section is intended to
apply and the section as a whole is intended to apply in other
circumstances.
It is not the purpose of this section to induce you to infringe any
patents or other property right claims or to contest validity of any
such claims; this section has the sole purpose of protecting the
integrity of the free software distribution system, which is
implemented by public license practices. Many people have made
generous contributions to the wide range of software distributed
through that system in reliance on consistent application of that
system; it is up to the author/donor to decide if he or she is willing
to distribute software through any other system and a licensee cannot
impose that choice.
This section is intended to make thoroughly clear what is believed to
be a consequence of the rest of this License.
8. If the distribution and/or use of the Program is restricted in
certain countries either by patents or by copyrighted interfaces, the
original copyright holder who places the Program under this License
may add an explicit geographical distribution limitation excluding
those countries, so that distribution is permitted only in or among
countries not thus excluded. In such case, this License incorporates
the limitation as if written in the body of this License.
9. The Free Software Foundation may publish revised and/or new versions
of the General Public License from time to time. Such new versions will
be similar in spirit to the present version, but may differ in detail to
address new problems or concerns.
Each version is given a distinguishing version number. If the Program
specifies a version number of this License which applies to it and "any
later version", you have the option of following the terms and conditions
either of that version or of any later version published by the Free
Software Foundation. If the Program does not specify a version number of
this License, you may choose any version ever published by the Free Software
Foundation.
10. If you wish to incorporate parts of the Program into other free
programs whose distribution conditions are different, write to the author
to ask for permission. For software which is copyrighted by the Free
Software Foundation, write to the Free Software Foundation; we sometimes
make exceptions for this. Our decision will be guided by the two goals
of preserving the free status of all derivatives of our free software and
of promoting the sharing and reuse of software generally.
NO WARRANTY
11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
REPAIR OR CORRECTION.
12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES.
END OF TERMS AND CONDITIONS
How to Apply These Terms to Your New Programs
If you develop a new program, and you want it to be of the greatest
possible use to the public, the best way to achieve this is to make it
free software which everyone can redistribute and change under these terms.
To do so, attach the following notices to the program. It is safest
to attach them to the start of each source file to most effectively
convey the exclusion of warranty; and each file should have at least
the "copyright" line and a pointer to where the full notice is found.
<one line to give the program's name and a brief idea of what it does.>
Copyright (C) <year> <name of author>
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License along
with this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
Also add information on how to contact you by electronic and paper mail.
If the program is interactive, make it output a short notice like this
when it starts in an interactive mode:
Gnomovision version 69, Copyright (C) year name of author
Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
This is free software, and you are welcome to redistribute it
under certain conditions; type `show c' for details.
The hypothetical commands `show w' and `show c' should show the appropriate
parts of the General Public License. Of course, the commands you use may
be called something other than `show w' and `show c'; they could even be
mouse-clicks or menu items--whatever suits your program.
You should also get your employer (if you work as a programmer) or your
school, if any, to sign a "copyright disclaimer" for the program, if
necessary. Here is a sample; alter the names:
Yoyodyne, Inc., hereby disclaims all copyright interest in the program
`Gnomovision' (which makes passes at compilers) written by James Hacker.
<signature of Ty Coon>, 1 April 1989
Ty Coon, President of Vice
This General Public License does not permit incorporating your program into
proprietary programs. If your program is a subroutine library, you may
consider it more useful to permit linking proprietary applications with the
library. If this is what you want to do, use the GNU Lesser General
Public License instead of this License.

30
raytracer/nvpro_core/third_party/zstd/LICENSE vendored Normal file
View file

@ -0,0 +1,30 @@
BSD License
For Zstandard software
Copyright (c) Meta Platforms, Inc. and affiliates. All rights reserved.
Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
* Neither the name Facebook, nor Meta, nor the names of its contributors may
be used to endorse or promote products derived from this software without
specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR
ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

223
raytracer/nvpro_core/third_party/zstd/README.md vendored Normal file
View file

@ -0,0 +1,223 @@
<p align="center"><img src="https://raw.githubusercontent.com/facebook/zstd/dev/doc/images/zstd_logo86.png" alt="Zstandard"></p>
__Zstandard__, or `zstd` as short version, is a fast lossless compression algorithm,
targeting real-time compression scenarios at zlib-level and better compression ratios.
It's backed by a very fast entropy stage, provided by [Huff0 and FSE library](https://github.com/Cyan4973/FiniteStateEntropy).
Zstandard's format is stable and documented in [RFC8878](https://datatracker.ietf.org/doc/html/rfc8878). Multiple independent implementations are already available.
This repository represents the reference implementation, provided as an open-source dual [BSD](LICENSE) and [GPLv2](COPYING) licensed **C** library,
and a command line utility producing and decoding `.zst`, `.gz`, `.xz` and `.lz4` files.
Should your project require another programming language,
a list of known ports and bindings is provided on [Zstandard homepage](https://facebook.github.io/zstd/#other-languages).
**Development branch status:**
[![Build Status][travisDevBadge]][travisLink]
[![Build status][CircleDevBadge]][CircleLink]
[![Build status][CirrusDevBadge]][CirrusLink]
[![Fuzzing Status][OSSFuzzBadge]][OSSFuzzLink]
[travisDevBadge]: https://api.travis-ci.com/facebook/zstd.svg?branch=dev "Continuous Integration test suite"
[travisLink]: https://travis-ci.com/facebook/zstd
[CircleDevBadge]: https://circleci.com/gh/facebook/zstd/tree/dev.svg?style=shield "Short test suite"
[CircleLink]: https://circleci.com/gh/facebook/zstd
[CirrusDevBadge]: https://api.cirrus-ci.com/github/facebook/zstd.svg?branch=dev
[CirrusLink]: https://cirrus-ci.com/github/facebook/zstd
[OSSFuzzBadge]: https://oss-fuzz-build-logs.storage.googleapis.com/badges/zstd.svg
[OSSFuzzLink]: https://bugs.chromium.org/p/oss-fuzz/issues/list?sort=-opened&can=1&q=proj:zstd
## Benchmarks
For reference, several fast compression algorithms were tested and compared
on a desktop running Ubuntu 20.04 (`Linux 5.11.0-41-generic`),
with a Core i7-9700K CPU @ 4.9GHz,
using [lzbench], an open-source in-memory benchmark by @inikep
compiled with [gcc] 9.3.0,
on the [Silesia compression corpus].
[lzbench]: https://github.com/inikep/lzbench
[Silesia compression corpus]: https://sun.aei.polsl.pl//~sdeor/index.php?page=silesia
[gcc]: https://gcc.gnu.org/
| Compressor name | Ratio | Compression| Decompress.|
| --------------- | ------| -----------| ---------- |
| **zstd 1.5.1 -1** | 2.887 | 530 MB/s | 1700 MB/s |
| [zlib] 1.2.11 -1 | 2.743 | 95 MB/s | 400 MB/s |
| brotli 1.0.9 -0 | 2.702 | 395 MB/s | 450 MB/s |
| **zstd 1.5.1 --fast=1** | 2.437 | 600 MB/s | 2150 MB/s |
| **zstd 1.5.1 --fast=3** | 2.239 | 670 MB/s | 2250 MB/s |
| quicklz 1.5.0 -1 | 2.238 | 540 MB/s | 760 MB/s |
| **zstd 1.5.1 --fast=4** | 2.148 | 710 MB/s | 2300 MB/s |
| lzo1x 2.10 -1 | 2.106 | 660 MB/s | 845 MB/s |
| [lz4] 1.9.3 | 2.101 | 740 MB/s | 4500 MB/s |
| lzf 3.6 -1 | 2.077 | 410 MB/s | 830 MB/s |
| snappy 1.1.9 | 2.073 | 550 MB/s | 1750 MB/s |
[zlib]: https://www.zlib.net/
[lz4]: https://lz4.github.io/lz4/
The negative compression levels, specified with `--fast=#`,
offer faster compression and decompression speed
at the cost of compression ratio (compared to level 1).
Zstd can also offer stronger compression ratios at the cost of compression speed.
Speed vs Compression trade-off is configurable by small increments.
Decompression speed is preserved and remains roughly the same at all settings,
a property shared by most LZ compression algorithms, such as [zlib] or lzma.
The following tests were run
on a server running Linux Debian (`Linux version 4.14.0-3-amd64`)
with a Core i7-6700K CPU @ 4.0GHz,
using [lzbench], an open-source in-memory benchmark by @inikep
compiled with [gcc] 7.3.0,
on the [Silesia compression corpus].
Compression Speed vs Ratio | Decompression Speed
---------------------------|--------------------
![Compression Speed vs Ratio](doc/images/CSpeed2.png "Compression Speed vs Ratio") | ![Decompression Speed](doc/images/DSpeed3.png "Decompression Speed")
A few other algorithms can produce higher compression ratios at slower speeds, falling outside of the graph.
For a larger picture including slow modes, [click on this link](doc/images/DCspeed5.png).
## The case for Small Data compression
Previous charts provide results applicable to typical file and stream scenarios (several MB). Small data comes with different perspectives.
The smaller the amount of data to compress, the more difficult it is to compress. This problem is common to all compression algorithms, and reason is, compression algorithms learn from past data how to compress future data. But at the beginning of a new data set, there is no "past" to build upon.
To solve this situation, Zstd offers a __training mode__, which can be used to tune the algorithm for a selected type of data.
Training Zstandard is achieved by providing it with a few samples (one file per sample). The result of this training is stored in a file called "dictionary", which must be loaded before compression and decompression.
Using this dictionary, the compression ratio achievable on small data improves dramatically.
The following example uses the `github-users` [sample set](https://github.com/facebook/zstd/releases/tag/v1.1.3), created from [github public API](https://developer.github.com/v3/users/#get-all-users).
It consists of roughly 10K records weighing about 1KB each.
Compression Ratio | Compression Speed | Decompression Speed
------------------|-------------------|--------------------
![Compression Ratio](doc/images/dict-cr.png "Compression Ratio") | ![Compression Speed](doc/images/dict-cs.png "Compression Speed") | ![Decompression Speed](doc/images/dict-ds.png "Decompression Speed")
These compression gains are achieved while simultaneously providing _faster_ compression and decompression speeds.
Training works if there is some correlation in a family of small data samples. The more data-specific a dictionary is, the more efficient it is (there is no _universal dictionary_).
Hence, deploying one dictionary per type of data will provide the greatest benefits.
Dictionary gains are mostly effective in the first few KB. Then, the compression algorithm will gradually use previously decoded content to better compress the rest of the file.
### Dictionary compression How To:
1. Create the dictionary
`zstd --train FullPathToTrainingSet/* -o dictionaryName`
2. Compress with dictionary
`zstd -D dictionaryName FILE`
3. Decompress with dictionary
`zstd -D dictionaryName --decompress FILE.zst`
## Build instructions
`make` is the officially maintained build system of this project.
All other build systems are "compatible" and 3rd-party maintained,
they may feature small differences in advanced options.
When your system allows it, prefer using `make` to build `zstd` and `libzstd`.
### Makefile
If your system is compatible with standard `make` (or `gmake`),
invoking `make` in root directory will generate `zstd` cli in root directory.
It will also create `libzstd` into `lib/`.
Other available options include:
- `make install` : create and install zstd cli, library and man pages
- `make check` : create and run `zstd`, test its behavior on local platform
The `Makefile` follows the [GNU Standard Makefile conventions](https://www.gnu.org/prep/standards/html_node/Makefile-Conventions.html),
allowing staged install, standard flags, directory variables and command variables.
For advanced use cases, specialized compilation flags which control binary generation
are documented in [`lib/README.md`](lib/README.md#modular-build) for the `libzstd` library
and in [`programs/README.md`](programs/README.md#compilation-variables) for the `zstd` CLI.
### cmake
A `cmake` project generator is provided within `build/cmake`.
It can generate Makefiles or other build scripts
to create `zstd` binary, and `libzstd` dynamic and static libraries.
By default, `CMAKE_BUILD_TYPE` is set to `Release`.
#### Support for Fat (Universal2) Output
`zstd` can be built and installed with support for both Apple Silicon (M1/M2) as well as Intel by using CMake's Universal2 support.
To perform a Fat/Universal2 build and install use the following commands:
```bash
cmake -B build-cmake-debug -S build/cmake -G Ninja -DCMAKE_OSX_ARCHITECTURES="x86_64;x86_64h;arm64"
cd build-cmake-debug
ninja
sudo ninja install
```
### Meson
A Meson project is provided within [`build/meson`](build/meson). Follow
build instructions in that directory.
You can also take a look at [`.travis.yml`](.travis.yml) file for an
example about how Meson is used to build this project.
Note that default build type is **release**.
### VCPKG
You can build and install zstd [vcpkg](https://github.com/Microsoft/vcpkg/) dependency manager:
git clone https://github.com/Microsoft/vcpkg.git
cd vcpkg
./bootstrap-vcpkg.sh
./vcpkg integrate install
./vcpkg install zstd
The zstd port in vcpkg is kept up to date by Microsoft team members and community contributors.
If the version is out of date, please [create an issue or pull request](https://github.com/Microsoft/vcpkg) on the vcpkg repository.
### Visual Studio (Windows)
Going into `build` directory, you will find additional possibilities:
- Projects for Visual Studio 2005, 2008 and 2010.
+ VS2010 project is compatible with VS2012, VS2013, VS2015 and VS2017.
- Automated build scripts for Visual compiler by [@KrzysFR](https://github.com/KrzysFR), in `build/VS_scripts`,
which will build `zstd` cli and `libzstd` library without any need to open Visual Studio solution.
### Buck
You can build the zstd binary via buck by executing: `buck build programs:zstd` from the root of the repo.
The output binary will be in `buck-out/gen/programs/`.
## Testing
You can run quick local smoke tests by running `make check`.
If you can't use `make`, execute the `playTest.sh` script from the `src/tests` directory.
Two env variables `$ZSTD_BIN` and `$DATAGEN_BIN` are needed for the test script to locate the `zstd` and `datagen` binary.
For information on CI testing, please refer to `TESTING.md`.
## Status
Zstandard is currently deployed within Facebook and many other large cloud infrastructures.
It is run continuously to compress large amounts of data in multiple formats and use cases.
Zstandard is considered safe for production environments.
## License
Zstandard is dual-licensed under [BSD](LICENSE) and [GPLv2](COPYING).
## Contributing
The `dev` branch is the one where all contributions are merged before reaching `release`.
If you plan to propose a patch, please commit into the `dev` branch, or its own feature branch.
Direct commit to `release` are not permitted.
For more information, please read [CONTRIBUTING](CONTRIBUTING.md).

43
raytracer/nvpro_core/third_party/zstd/TESTING.md vendored Normal file
View file

@ -0,0 +1,43 @@
Testing
=======
Zstandard CI testing is split up into three sections:
short, medium, and long tests.
Short Tests
-----------
Short tests run on CircleCI for new commits on every branch and pull request.
They consist of the following tests:
- Compilation on all supported targets (x86, x86_64, ARM, AArch64, PowerPC, and PowerPC64)
- Compilation on various versions of gcc, clang, and g++
- `tests/playTests.sh` on x86_64, without the tests on long data (CLI tests)
- Small tests (`tests/legacy.c`, `tests/longmatch.c`) on x64_64
Medium Tests
------------
Medium tests run on every commit and pull request to `dev` branch, on TravisCI.
They consist of the following tests:
- The following tests run with UBsan and Asan on x86_64 and x86, as well as with
Msan on x86_64
- `tests/playTests.sh --test-large-data`
- Fuzzer tests: `tests/fuzzer.c`, `tests/zstreamtest.c`, and `tests/decodecorpus.c`
- `tests/zstreamtest.c` under Tsan (streaming mode, including multithreaded mode)
- Valgrind Test (`make -C tests test-valgrind`) (testing CLI and fuzzer under `valgrind`)
- Fuzzer tests (see above) on ARM, AArch64, PowerPC, and PowerPC64
Long Tests
----------
Long tests run on all commits to `release` branch,
and once a day on the current version of `dev` branch,
on TravisCI.
They consist of the following tests:
- Entire test suite (including fuzzers and some other specialized tests) on:
- x86_64 and x86 with UBsan and Asan
- x86_64 with Msan
- ARM, AArch64, PowerPC, and PowerPC64
- Streaming mode fuzzer with Tsan (for the `zstdmt` testing)
- ZlibWrapper tests, including under valgrind
- Versions test (ensuring `zstd` can decode files from all previous versions)
- `pzstd` with asan and tsan, as well as in 32-bits mode
- Testing `zstd` with legacy mode off
- Entire test suite and make install on macOS

0
raytracer/nvpro_core/third_party/zstd/build/LICENSE vendored Normal file
View file

56
raytracer/nvpro_core/third_party/zstd/build/README.md vendored Normal file
View file

@ -0,0 +1,56 @@
Projects for various integrated development environments (IDE)
==============================================================
#### Included projects
The following projects are included with the zstd distribution:
- `cmake` - CMake project contributed by Artyom Dymchenko
- `VS2005` - Visual Studio 2005 Project (this project has been moved to the contrib directory and will no longer be supported)
- `VS2008` - Visual Studio 2008 project
- `VS2010` - Visual Studio 2010 project (which also works well with Visual Studio 2012, 2013, 2015)
- `VS_scripts` - command line scripts prepared for Visual Studio compilation without IDE
#### How to compile zstd with Visual Studio
1. Install Visual Studio e.g. VS 2015 Community Edition (it's free).
2. Download the latest version of zstd from https://github.com/facebook/zstd/releases
3. Decompress ZIP archive.
4. Go to decompressed directory then to `projects` then `VS2010` and open `zstd.sln`
5. Visual Studio will ask about converting VS2010 project to VS2015 and you should agree.
6. Change `Debug` to `Release` and if you have 64-bit Windows change also `Win32` to `x64`.
7. Press F7 on keyboard or select `BUILD` from the menu bar and choose `Build Solution`.
8. If compilation will be fine a compiled executable will be in `projects\VS2010\bin\x64\Release\zstd.exe`
#### Projects available within zstd.sln
The Visual Studio solution file `visual\VS2010\zstd.sln` contains many projects that will be compiled to the
`visual\VS2010\bin\$(Platform)_$(Configuration)` directory. For example `zstd` set to `x64` and
`Release` will be compiled to `visual\VS2010\bin\x64_Release\zstd.exe`. The solution file contains the
following projects:
- `zstd` : Command Line Utility, supporting gzip-like arguments
- `datagen` : Synthetic and parametrable data generator, for tests
- `fullbench` : Precisely measure speed for each zstd inner functions
- `fuzzer` : Test tool, to check zstd integrity on target platform
- `libzstd` : A static ZSTD library compiled to `libzstd_static.lib`
- `libzstd-dll` : A dynamic ZSTD library (DLL) compiled to `libzstd.dll` with the import library `libzstd.lib`
- `fullbench-dll` : The fullbench program compiled with the import library; the executable requires ZSTD DLL
#### Using ZSTD DLL with Microsoft Visual C++ project
The header file `lib\zstd.h` and the import library
`visual\VS2010\bin\$(Platform)_$(Configuration)\libzstd.lib` are required to compile
a project using Visual C++.
1. The path to header files should be added to `Additional Include Directories` that can
be found in Project Properties of Visual Studio IDE in the `C/C++` Property Pages on the `General` page.
2. The import library has to be added to `Additional Dependencies` that can
be found in Project Properties in the `Linker` Property Pages on the `Input` page.
If one will provide only the name `libzstd.lib` without a full path to the library
then the directory has to be added to `Linker\General\Additional Library Directories`.
The compiled executable will require ZSTD DLL which is available at
`visual\VS2010\bin\$(Platform)_$(Configuration)\libzstd.dll`.

212
raytracer/nvpro_core/third_party/zstd/build/cmake/CMakeLists.txt vendored Normal file
View file

@ -0,0 +1,212 @@
# ################################################################
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under both the BSD-style license (found in the
# LICENSE file in the root directory of this source tree) and the GPLv2 (found
# in the COPYING file in the root directory of this source tree).
# ################################################################
cmake_minimum_required(VERSION 2.8.12 FATAL_ERROR)
# As of 2018-12-26 ZSTD has been validated to build with cmake version 3.13.2 new policies.
# Set and use the newest cmake policies that are validated to work
set(ZSTD_MAX_VALIDATED_CMAKE_MAJOR_VERSION "3")
set(ZSTD_MAX_VALIDATED_CMAKE_MINOR_VERSION "13") #Policies never changed at PATCH level
if("${CMAKE_MAJOR_VERSION}" LESS 3)
set(ZSTD_CMAKE_POLICY_VERSION "${CMAKE_VERSION}")
elseif( "${ZSTD_MAX_VALIDATED_CMAKE_MAJOR_VERSION}" EQUAL "${CMAKE_MAJOR_VERSION}" AND
"${ZSTD_MAX_VALIDATED_CMAKE_MINOR_VERSION}" GREATER "${CMAKE_MINOR_VERSION}")
set(ZSTD_CMAKE_POLICY_VERSION "${CMAKE_VERSION}")
else()
set(ZSTD_CMAKE_POLICY_VERSION "${ZSTD_MAX_VALIDATED_CMAKE_MAJOR_VERSION}.${ZSTD_MAX_VALIDATED_CMAKE_MINOR_VERSION}.0")
endif()
cmake_policy(VERSION ${ZSTD_CMAKE_POLICY_VERSION})
set(CMAKE_BUILD_WITH_INSTALL_RPATH on)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/CMakeModules")
set(ZSTD_SOURCE_DIR "${CMAKE_CURRENT_SOURCE_DIR}/../..")
set(LIBRARY_DIR ${ZSTD_SOURCE_DIR}/lib)
# Parse version
include(GetZstdLibraryVersion)
GetZstdLibraryVersion(${LIBRARY_DIR}/zstd.h zstd_VERSION_MAJOR zstd_VERSION_MINOR zstd_VERSION_PATCH)
if( CMAKE_MAJOR_VERSION LESS 3 )
## Provide cmake 3+ behavior for older versions of cmake
project(zstd)
set(PROJECT_VERSION_MAJOR ${zstd_VERSION_MAJOR})
set(PROJECT_VERSION_MINOR ${zstd_VERSION_MINOR})
set(PROJECT_VERSION_PATCH ${zstd_VERSION_PATCH})
set(PROJECT_VERSION "${zstd_VERSION_MAJOR}.${zstd_VERSION_MINOR}.${zstd_VERSION_PATCH}")
enable_language(C) # Main library is in C
enable_language(ASM) # And ASM
enable_language(CXX) # Testing contributed code also utilizes CXX
else()
project(zstd
VERSION "${zstd_VERSION_MAJOR}.${zstd_VERSION_MINOR}.${zstd_VERSION_PATCH}"
LANGUAGES C # Main library is in C
ASM # And ASM
CXX # Testing contributed code also utilizes CXX
)
endif()
message(STATUS "ZSTD VERSION: ${zstd_VERSION}")
set(zstd_HOMEPAGE_URL "https://facebook.github.io/zstd")
set(zstd_DESCRIPTION "Zstandard is a real-time compression algorithm, providing high compression ratios.")
# Set a default build type if none was specified
if(NOT CMAKE_BUILD_TYPE AND NOT CMAKE_CONFIGURATION_TYPES)
message(STATUS "Setting build type to 'Release' as none was specified.")
set(CMAKE_BUILD_TYPE Release CACHE STRING "Choose the type of build." FORCE)
# Set the possible values of build type for cmake-gui
set_property(CACHE CMAKE_BUILD_TYPE PROPERTY STRINGS "Debug" "Release" "MinSizeRel" "RelWithDebInfo")
endif()
include(GNUInstallDirs)
#-----------------------------------------------------------------------------
# Add extra compilation flags
#-----------------------------------------------------------------------------
include(AddZstdCompilationFlags)
ADD_ZSTD_COMPILATION_FLAGS()
# Always hide XXHash symbols
add_definitions(-DXXH_NAMESPACE=ZSTD_)
#-----------------------------------------------------------------------------
# Installation variables
#-----------------------------------------------------------------------------
message(STATUS "CMAKE_INSTALL_PREFIX: ${CMAKE_INSTALL_PREFIX}")
message(STATUS "CMAKE_INSTALL_LIBDIR: ${CMAKE_INSTALL_LIBDIR}")
#-----------------------------------------------------------------------------
# Options
#-----------------------------------------------------------------------------
# Legacy support
option(ZSTD_LEGACY_SUPPORT "LEGACY SUPPORT" ON)
if (ZSTD_LEGACY_SUPPORT)
message(STATUS "ZSTD_LEGACY_SUPPORT defined!")
set(ZSTD_LEGACY_LEVEL 5 CACHE STRING "")
add_definitions(-DZSTD_LEGACY_SUPPORT=${ZSTD_LEGACY_LEVEL})
else ()
message(STATUS "ZSTD_LEGACY_SUPPORT not defined!")
add_definitions(-DZSTD_LEGACY_SUPPORT=0)
endif ()
if (ANDROID)
set(ZSTD_MULTITHREAD_SUPPORT_DEFAULT OFF)
else()
set(ZSTD_MULTITHREAD_SUPPORT_DEFAULT ON)
endif()
# Multi-threading support
option(ZSTD_MULTITHREAD_SUPPORT "MULTITHREADING SUPPORT" ${ZSTD_MULTITHREAD_SUPPORT_DEFAULT})
if (ZSTD_MULTITHREAD_SUPPORT)
message(STATUS "ZSTD_MULTITHREAD_SUPPORT is enabled")
else ()
message(STATUS "ZSTD_MULTITHREAD_SUPPORT is disabled")
endif ()
option(ZSTD_BUILD_PROGRAMS "BUILD PROGRAMS" ON)
option(ZSTD_BUILD_CONTRIB "BUILD CONTRIB" OFF)
# Respect the conventional CMake option for enabling tests if it was specified on the first configure
if (BUILD_TESTING)
set(ZSTD_BUILD_TESTS_default ON)
else()
set(ZSTD_BUILD_TESTS_default OFF)
endif()
option(ZSTD_BUILD_TESTS "BUILD TESTS" ${ZSTD_BUILD_TESTS_default})
if (MSVC)
option(ZSTD_USE_STATIC_RUNTIME "LINK TO STATIC RUN-TIME LIBRARIES" OFF)
endif ()
#-----------------------------------------------------------------------------
# External dependencies
#-----------------------------------------------------------------------------
if (ZSTD_MULTITHREAD_SUPPORT AND UNIX)
set(THREADS_PREFER_PTHREAD_FLAG ON)
find_package(Threads REQUIRED)
if(CMAKE_USE_PTHREADS_INIT)
set(THREADS_LIBS "${CMAKE_THREAD_LIBS_INIT}")
else()
message(SEND_ERROR "ZSTD currently does not support thread libraries other than pthreads")
endif()
endif ()
#-----------------------------------------------------------------------------
# Add source directories
#-----------------------------------------------------------------------------
add_subdirectory(lib)
option(ZSTD_PROGRAMS_LINK_SHARED "PROGRAMS LINK SHARED" OFF)
if (ZSTD_BUILD_PROGRAMS)
if (NOT ZSTD_BUILD_STATIC AND NOT ZSTD_PROGRAMS_LINK_SHARED)
message(SEND_ERROR "You need to build static library to build zstd CLI")
elseif(NOT ZSTD_BUILD_SHARED AND ZSTD_PROGRAMS_LINK_SHARED)
message(SEND_ERROR "You need to build shared library to build zstd CLI")
endif ()
add_subdirectory(programs)
endif ()
if (ZSTD_BUILD_TESTS)
enable_testing()
if (NOT ZSTD_BUILD_STATIC)
message(SEND_ERROR "You need to build static library to build tests")
endif ()
add_subdirectory(tests)
endif ()
if (ZSTD_BUILD_CONTRIB)
add_subdirectory(contrib)
endif ()
#-----------------------------------------------------------------------------
# Add clean-all target
#-----------------------------------------------------------------------------
add_custom_target(clean-all
COMMAND ${CMAKE_BUILD_TOOL} clean
COMMAND rm -rf ${CMAKE_BINARY_DIR}/
)
#-----------------------------------------------------------------------------
# Generate Package Config files
#
# This section is based on the boiler plate code from:
# https://cmake.org/cmake/help/latest/manual/cmake-packages.7.html#creating-packages
#-----------------------------------------------------------------------------
include(CMakePackageConfigHelpers)
write_basic_package_version_file(
"${CMAKE_CURRENT_BINARY_DIR}/zstdConfigVersion.cmake"
VERSION ${zstd_VERSION}
COMPATIBILITY SameMajorVersion
)
# A Package Config file that works from the build directory
export(EXPORT zstdExports
FILE "${CMAKE_CURRENT_BINARY_DIR}/zstdTargets.cmake"
NAMESPACE zstd::
)
configure_file(zstdConfig.cmake
"${CMAKE_CURRENT_BINARY_DIR}/zstdConfig.cmake"
COPYONLY
)
# A Package Config file that works from the installation directory
set(ConfigPackageLocation ${CMAKE_INSTALL_LIBDIR}/cmake/zstd)
install(EXPORT zstdExports
FILE zstdTargets.cmake
NAMESPACE zstd::
DESTINATION ${ConfigPackageLocation}
)
install(FILES
zstdConfig.cmake
"${CMAKE_CURRENT_BINARY_DIR}/zstdConfigVersion.cmake"
DESTINATION ${ConfigPackageLocation}
)

121
raytracer/nvpro_core/third_party/zstd/build/cmake/CMakeModules/AddZstdCompilationFlags.cmake vendored Normal file
View file

@ -0,0 +1,121 @@
include(CheckCXXCompilerFlag)
include(CheckCCompilerFlag)
# VERSION_GREATER_EQUAL requires CMake 3.7 or later.
# https://cmake.org/cmake/help/latest/command/if.html#version-greater-equal
if (CMAKE_VERSION VERSION_LESS 3.18)
set(ZSTD_HAVE_CHECK_LINKER_FLAG false)
else ()
set(ZSTD_HAVE_CHECK_LINKER_FLAG true)
endif ()
if (ZSTD_HAVE_CHECK_LINKER_FLAG)
include(CheckLinkerFlag)
endif()
function(EnableCompilerFlag _flag _C _CXX _LD)
string(REGEX REPLACE "\\+" "PLUS" varname "${_flag}")
string(REGEX REPLACE "[^A-Za-z0-9]+" "_" varname "${varname}")
string(REGEX REPLACE "^_+" "" varname "${varname}")
string(TOUPPER "${varname}" varname)
if (_C)
CHECK_C_COMPILER_FLAG(${_flag} C_FLAG_${varname})
if (C_FLAG_${varname})
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${_flag}" PARENT_SCOPE)
endif ()
endif ()
if (_CXX)
CHECK_CXX_COMPILER_FLAG(${_flag} CXX_FLAG_${varname})
if (CXX_FLAG_${varname})
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${_flag}" PARENT_SCOPE)
endif ()
endif ()
if (_LD)
# We never add a linker flag with CMake < 3.18. We will
# implement CHECK_LINKER_FLAG() like feature for CMake < 3.18
# or require CMake >= 3.18 when we need to add a required
# linker flag in future.
#
# We also skip linker flags check for MSVC compilers (which includes
# clang-cl) since currently check_linker_flag() doesn't give correct
# results for this configuration,
# see: https://gitlab.kitware.com/cmake/cmake/-/issues/22023
if (ZSTD_HAVE_CHECK_LINKER_FLAG AND NOT MSVC)
CHECK_LINKER_FLAG(C ${_flag} LD_FLAG_${varname})
else ()
set(LD_FLAG_${varname} false)
endif ()
if (LD_FLAG_${varname})
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} ${_flag}" PARENT_SCOPE)
set(CMAKE_SHARED_LINKER_FLAGS "${CMAKE_SHARED_LINKER_FLAGS} ${_flag}" PARENT_SCOPE)
endif ()
endif ()
endfunction()
macro(ADD_ZSTD_COMPILATION_FLAGS)
if (CMAKE_CXX_COMPILER_ID MATCHES "GNU|Clang" OR MINGW) #Not only UNIX but also WIN32 for MinGW
# It's possible to select the exact standard used for compilation.
# It's not necessary, but can be employed for specific purposes.
# Note that zstd source code is compatible with both C++98 and above
# and C-gnu90 (c90 + long long + variadic macros ) and above
# EnableCompilerFlag("-std=c++11" false true) # Set C++ compilation to c++11 standard
# EnableCompilerFlag("-std=c99" true false) # Set C compiation to c99 standard
if (CMAKE_CXX_COMPILER_ID MATCHES "Clang" AND MSVC)
# clang-cl normally maps -Wall to -Weverything.
EnableCompilerFlag("/clang:-Wall" true true false)
else ()
EnableCompilerFlag("-Wall" true true false)
endif ()
EnableCompilerFlag("-Wextra" true true false)
EnableCompilerFlag("-Wundef" true true false)
EnableCompilerFlag("-Wshadow" true true false)
EnableCompilerFlag("-Wcast-align" true true false)
EnableCompilerFlag("-Wcast-qual" true true false)
EnableCompilerFlag("-Wstrict-prototypes" true false false)
# Enable asserts in Debug mode
if (CMAKE_BUILD_TYPE MATCHES "Debug")
EnableCompilerFlag("-DDEBUGLEVEL=1" true true false)
endif ()
# Add noexecstack flags
# LDFLAGS
EnableCompilerFlag("-z noexecstack" false false true)
# CFLAGS & CXXFLAGS
EnableCompilerFlag("-Qunused-arguments" true true false)
EnableCompilerFlag("-Wa,--noexecstack" true true false)
elseif (MSVC) # Add specific compilation flags for Windows Visual
set(ACTIVATE_MULTITHREADED_COMPILATION "ON" CACHE BOOL "activate multi-threaded compilation (/MP flag)")
if (CMAKE_GENERATOR MATCHES "Visual Studio" AND ACTIVATE_MULTITHREADED_COMPILATION)
EnableCompilerFlag("/MP" true true false)
endif ()
# UNICODE SUPPORT
EnableCompilerFlag("/D_UNICODE" true true false)
EnableCompilerFlag("/DUNICODE" true true false)
# Enable asserts in Debug mode
if (CMAKE_BUILD_TYPE MATCHES "Debug")
EnableCompilerFlag("/DDEBUGLEVEL=1" true true false)
endif ()
endif ()
# Remove duplicates compilation flags
foreach (flag_var CMAKE_C_FLAGS CMAKE_C_FLAGS_DEBUG CMAKE_C_FLAGS_RELEASE
CMAKE_C_FLAGS_MINSIZEREL CMAKE_C_FLAGS_RELWITHDEBINFO
CMAKE_CXX_FLAGS CMAKE_CXX_FLAGS_DEBUG CMAKE_CXX_FLAGS_RELEASE
CMAKE_CXX_FLAGS_MINSIZEREL CMAKE_CXX_FLAGS_RELWITHDEBINFO)
if( ${flag_var} )
separate_arguments(${flag_var})
string(REPLACE ";" " " ${flag_var} "${${flag_var}}")
endif()
endforeach ()
if (MSVC AND ZSTD_USE_STATIC_RUNTIME)
foreach (flag_var CMAKE_C_FLAGS CMAKE_C_FLAGS_DEBUG CMAKE_C_FLAGS_RELEASE
CMAKE_C_FLAGS_MINSIZEREL CMAKE_C_FLAGS_RELWITHDEBINFO
CMAKE_CXX_FLAGS CMAKE_CXX_FLAGS_DEBUG CMAKE_CXX_FLAGS_RELEASE
CMAKE_CXX_FLAGS_MINSIZEREL CMAKE_CXX_FLAGS_RELWITHDEBINFO)
if ( ${flag_var} )
string(REGEX REPLACE "/MD" "/MT" ${flag_var} "${${flag_var}}")
endif()
endforeach ()
endif ()
endmacro()

49
raytracer/nvpro_core/third_party/zstd/build/cmake/CMakeModules/FindLibLZ4.cmake vendored Normal file
View file

@ -0,0 +1,49 @@
# Find LibLZ4
#
# Find LibLZ4 headers and library
#
# Result Variables
#
# LIBLZ4_FOUND - True if lz4 is found
# LIBLZ4_INCLUDE_DIRS - lz4 headers directories
# LIBLZ4_LIBRARIES - lz4 libraries
# LIBLZ4_VERSION_MAJOR - The major version of lz4
# LIBLZ4_VERSION_MINOR - The minor version of lz4
# LIBLZ4_VERSION_RELEASE - The release version of lz4
# LIBLZ4_VERSION_STRING - version number string (e.g. 1.8.3)
#
# Hints
#
# Set ``LZ4_ROOT_DIR`` to the directory of lz4.h and lz4 library
set(_LIBLZ4_ROOT_HINTS
ENV LZ4_ROOT_DIR)
find_path( LIBLZ4_INCLUDE_DIR lz4.h
HINTS ${_LIBLZ4_ROOT_HINTS})
find_library( LIBLZ4_LIBRARY NAMES lz4 liblz4 liblz4_static
HINTS ${_LIBLZ4_ROOT_HINTS})
if(LIBLZ4_INCLUDE_DIR)
file(STRINGS "${LIBLZ4_INCLUDE_DIR}/lz4.h" LIBLZ4_HEADER_CONTENT REGEX "#define LZ4_VERSION_[A-Z]+ +[0-9]+")
string(REGEX REPLACE ".*#define LZ4_VERSION_MAJOR +([0-9]+).*" "\\1" LIBLZ4_VERSION_MAJOR "${LIBLZ4_HEADER_CONTENT}")
string(REGEX REPLACE ".*#define LZ4_VERSION_MINOR +([0-9]+).*" "\\1" LIBLZ4_VERSION_MINOR "${LIBLZ4_HEADER_CONTENT}")
string(REGEX REPLACE ".*#define LZ4_VERSION_RELEASE +([0-9]+).*" "\\1" LIBLZ4_VERSION_RELEASE "${LIBLZ4_HEADER_CONTENT}")
set(LIBLZ4_VERSION_STRING "${LIBLZ4_VERSION_MAJOR}.${LIBLZ4_VERSION_MINOR}.${LIBLZ4_VERSION_RELEASE}")
unset(LIBLZ4_HEADER_CONTENT)
endif()
include(FindPackageHandleStandardArgs)
FIND_PACKAGE_HANDLE_STANDARD_ARGS(LibLZ4 REQUIRED_VARS LIBLZ4_INCLUDE_DIR
LIBLZ4_LIBRARY
VERSION_VAR LIBLZ4_VERSION_STRING
FAIL_MESSAGE "Could NOT find LZ4, try to set the paths to lz4.h and lz4 library in environment variable LZ4_ROOT_DIR")
if (LIBLZ4_FOUND)
set(LIBLZ4_LIBRARIES ${LIBLZ4_LIBRARY})
set(LIBLZ4_INCLUDE_DIRS ${LIBLZ4_INCLUDE_DIR})
endif ()
mark_as_advanced( LIBLZ4_INCLUDE_DIR LIBLZ4_LIBRARY )

10
raytracer/nvpro_core/third_party/zstd/build/cmake/CMakeModules/GetZstdLibraryVersion.cmake vendored Normal file
View file

@ -0,0 +1,10 @@
function(GetZstdLibraryVersion _header _major _minor _patch)
# Read file content
file(READ ${_header} CONTENT)
string(REGEX MATCH ".*define ZSTD_VERSION_MAJOR *([0-9]+).*define ZSTD_VERSION_MINOR *([0-9]+).*define ZSTD_VERSION_RELEASE *([0-9]+)" VERSION_REGEX "${CONTENT}")
set(${_major} ${CMAKE_MATCH_1} PARENT_SCOPE)
set(${_minor} ${CMAKE_MATCH_2} PARENT_SCOPE)
set(${_patch} ${CMAKE_MATCH_3} PARENT_SCOPE)
endfunction()

23
raytracer/nvpro_core/third_party/zstd/build/cmake/CMakeModules/JoinPaths.cmake vendored Normal file
View file

@ -0,0 +1,23 @@
# This module provides function for joining paths
# known from most languages
#
# SPDX-License-Identifier: (MIT OR CC0-1.0)
# Copyright 2020 Jan Tojnar
# https://github.com/jtojnar/cmake-snips
#
# Modelled after Python’s os.path.join
# https://docs.python.org/3.7/library/os.path.html#os.path.join
# Windows not supported
function(join_paths joined_path first_path_segment)
set(temp_path "${first_path_segment}")
foreach(current_segment IN LISTS ARGN)
if(NOT ("${current_segment}" STREQUAL ""))
if(IS_ABSOLUTE "${current_segment}")
set(temp_path "${current_segment}")
else()
set(temp_path "${temp_path}/${current_segment}")
endif()
endif()
endforeach()
set(${joined_path} "${temp_path}" PARENT_SCOPE)
endfunction()

104
raytracer/nvpro_core/third_party/zstd/build/cmake/README.md vendored Normal file
View file

@ -0,0 +1,104 @@
# Cmake contributions
Contributions to the cmake build configurations are welcome. Please
use case sensitivity that matches modern (i.e. cmake version 2.6 and above)
conventions of using lower-case for commands, and upper-case for
variables.
## How to build
As cmake doesn't support command like `cmake clean`, it's recommended to perform an "out of source build".
To do this, you can create a new directory and build in it:
```sh
cd build/cmake
mkdir builddir
cd builddir
cmake ..
make
```
Then you can clean all cmake caches by simply delete the new directory:
```sh
rm -rf build/cmake/builddir
```
And of course, you can directly build in build/cmake:
```sh
cd build/cmake
cmake
make
```
To show cmake build options, you can:
```sh
cd build/cmake/builddir
cmake -LH ..
```
Bool options can be set to `ON/OFF` with `-D[option]=[ON/OFF]`. You can configure cmake options like this:
```sh
cd build/cmake/builddir
cmake -DZSTD_BUILD_TESTS=ON -DZSTD_LEGACY_SUPPORT=OFF ..
make
```
### referring
[Looking for a 'cmake clean' command to clear up CMake output](https://stackoverflow.com/questions/9680420/looking-for-a-cmake-clean-command-to-clear-up-cmake-output)
## CMake Style Recommendations
### Indent all code correctly, i.e. the body of
* if/else/endif
* foreach/endforeach
* while/endwhile
* macro/endmacro
* function/endfunction
Use spaces for indenting, 2, 3 or 4 spaces preferably. Use the same amount of
spaces for indenting as is used in the rest of the file. Do not use tabs.
### Upper/lower casing
Most important: use consistent upper- or lowercasing within one file !
In general, the all-lowercase style is preferred.
So, this is recommended:
```
add_executable(foo foo.c)
```
These forms are discouraged
```
ADD_EXECUTABLE(bar bar.c)
Add_Executable(hello hello.c)
aDd_ExEcUtAbLe(blub blub.c)
```
### End commands
To make the code easier to read, use empty commands for endforeach(), endif(),
endfunction(), endmacro() and endwhile(). Also, use empty else() commands.
For example, do this:
```
if(FOOVAR)
some_command(...)
else()
another_command(...)
endif()
```
and not this:
```
if(BARVAR)
some_other_command(...)
endif(BARVAR)
```
### Other resources for best practices
https://cmake.org/cmake/help/latest/manual/cmake-developer.7.html#modules

13
raytracer/nvpro_core/third_party/zstd/build/cmake/contrib/CMakeLists.txt vendored Normal file
View file

@ -0,0 +1,13 @@
# ################################################################
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under both the BSD-style license (found in the
# LICENSE file in the root directory of this source tree) and the GPLv2 (found
# in the COPYING file in the root directory of this source tree).
# ################################################################
project(contrib)
add_subdirectory(pzstd)
add_subdirectory(gen_html)

30
raytracer/nvpro_core/third_party/zstd/build/cmake/contrib/gen_html/CMakeLists.txt vendored Normal file
View file

@ -0,0 +1,30 @@
# ################################################################
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under both the BSD-style license (found in the
# LICENSE file in the root directory of this source tree) and the GPLv2 (found
# in the COPYING file in the root directory of this source tree).
# ################################################################
project(gen_html)
include(GetZstdLibraryVersion)
set(CMAKE_INCLUDE_CURRENT_DIR TRUE)
# Define programs directory, where sources and header files are located
set(LIBRARY_DIR ${ZSTD_SOURCE_DIR}/lib)
set(PROGRAMS_DIR ${ZSTD_SOURCE_DIR}/programs)
set(GENHTML_DIR ${ZSTD_SOURCE_DIR}/contrib/gen_html)
set(GENHTML_BINARY ${PROJECT_BINARY_DIR}/gen_html${CMAKE_EXECUTABLE_SUFFIX})
include_directories(${PROGRAMS_DIR} ${LIBRARY_DIR} ${LIBRARY_DIR}/common ${GENHTML_DIR})
add_executable(gen_html ${GENHTML_DIR}/gen_html.cpp)
GetZstdLibraryVersion(${LIBRARY_DIR}/zstd.h VMAJOR VMINOR VRELEASE)
set(LIBVERSION "${VMAJOR}.${VMINOR}.${VRELEASE}")
add_custom_target(zstd_manual.html ALL
${GENHTML_BINARY} "${LIBVERSION}" "${LIBRARY_DIR}/zstd.h" "${PROJECT_BINARY_DIR}/zstd_manual.html"
DEPENDS gen_html COMMENT "Update zstd manual")
install(FILES "${PROJECT_BINARY_DIR}/zstd_manual.html" DESTINATION "${CMAKE_INSTALL_DOCDIR}")

38
raytracer/nvpro_core/third_party/zstd/build/cmake/contrib/pzstd/CMakeLists.txt vendored Normal file
View file

@ -0,0 +1,38 @@
# ################################################################
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under both the BSD-style license (found in the
# LICENSE file in the root directory of this source tree) and the GPLv2 (found
# in the COPYING file in the root directory of this source tree).
# ################################################################
project(pzstd)
set(CMAKE_INCLUDE_CURRENT_DIR TRUE)
# Define programs directory, where sources and header files are located
set(LIBRARY_DIR ${ZSTD_SOURCE_DIR}/lib)
set(PROGRAMS_DIR ${ZSTD_SOURCE_DIR}/programs)
set(PZSTD_DIR ${ZSTD_SOURCE_DIR}/contrib/pzstd)
include_directories(${PROGRAMS_DIR} ${LIBRARY_DIR} ${LIBRARY_DIR}/common ${PZSTD_DIR})
add_executable(pzstd ${PROGRAMS_DIR}/util.c ${PZSTD_DIR}/main.cpp ${PZSTD_DIR}/Options.cpp ${PZSTD_DIR}/Pzstd.cpp ${PZSTD_DIR}/SkippableFrame.cpp)
set_property(TARGET pzstd APPEND PROPERTY COMPILE_DEFINITIONS "NDEBUG")
set_property(TARGET pzstd APPEND PROPERTY COMPILE_OPTIONS "-Wno-shadow")
if (ZSTD_BUILD_SHARED)
set(ZSTD_LIB libzstd_shared)
else()
set(ZSTD_LIB libzstd_static)
endif()
set(THREADS_PREFER_PTHREAD_FLAG ON)
find_package(Threads REQUIRED)
if (CMAKE_USE_PTHREADS_INIT)
target_link_libraries(pzstd ${ZSTD_LIB} ${CMAKE_THREAD_LIBS_INIT})
else()
message(SEND_ERROR "ZSTD currently does not support thread libraries other than pthreads")
endif()
install(TARGETS pzstd RUNTIME DESTINATION "bin")

181
raytracer/nvpro_core/third_party/zstd/build/cmake/lib/CMakeLists.txt vendored Normal file
View file

@ -0,0 +1,181 @@
# ################################################################
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under both the BSD-style license (found in the
# LICENSE file in the root directory of this source tree) and the GPLv2 (found
# in the COPYING file in the root directory of this source tree).
# ################################################################
project(libzstd C ASM)
set(CMAKE_INCLUDE_CURRENT_DIR TRUE)
option(ZSTD_BUILD_STATIC "BUILD STATIC LIBRARIES" ON)
option(ZSTD_BUILD_SHARED "BUILD SHARED LIBRARIES" ON)
if(NOT ZSTD_BUILD_SHARED AND NOT ZSTD_BUILD_STATIC)
message(SEND_ERROR "You need to build at least one flavor of libzstd")
endif()
# Define library directory, where sources and header files are located
include_directories(${LIBRARY_DIR} ${LIBRARY_DIR}/common)
file(GLOB CommonSources ${LIBRARY_DIR}/common/*.c)
file(GLOB CompressSources ${LIBRARY_DIR}/compress/*.c)
if (MSVC)
file(GLOB DecompressSources ${LIBRARY_DIR}/decompress/*.c)
add_compile_options(-DZSTD_DISABLE_ASM)
else ()
file(GLOB DecompressSources ${LIBRARY_DIR}/decompress/*.c ${LIBRARY_DIR}/decompress/*.S)
endif ()
file(GLOB DictBuilderSources ${LIBRARY_DIR}/dictBuilder/*.c)
set(Sources
${CommonSources}
${CompressSources}
${DecompressSources}
${DictBuilderSources})
file(GLOB CommonHeaders ${LIBRARY_DIR}/common/*.h)
file(GLOB CompressHeaders ${LIBRARY_DIR}/compress/*.h)
file(GLOB DecompressHeaders ${LIBRARY_DIR}/decompress/*.h)
file(GLOB DictBuilderHeaders ${LIBRARY_DIR}/dictBuilder/*.h)
set(Headers
${LIBRARY_DIR}/zstd.h
${CommonHeaders}
${CompressHeaders}
${DecompressHeaders}
${DictBuilderHeaders})
if (ZSTD_LEGACY_SUPPORT)
set(LIBRARY_LEGACY_DIR ${LIBRARY_DIR}/legacy)
include_directories(${LIBRARY_LEGACY_DIR})
set(Sources ${Sources}
${LIBRARY_LEGACY_DIR}/zstd_v01.c
${LIBRARY_LEGACY_DIR}/zstd_v02.c
${LIBRARY_LEGACY_DIR}/zstd_v03.c
${LIBRARY_LEGACY_DIR}/zstd_v04.c
${LIBRARY_LEGACY_DIR}/zstd_v05.c
${LIBRARY_LEGACY_DIR}/zstd_v06.c
${LIBRARY_LEGACY_DIR}/zstd_v07.c)
set(Headers ${Headers}
${LIBRARY_LEGACY_DIR}/zstd_legacy.h
${LIBRARY_LEGACY_DIR}/zstd_v01.h
${LIBRARY_LEGACY_DIR}/zstd_v02.h
${LIBRARY_LEGACY_DIR}/zstd_v03.h
${LIBRARY_LEGACY_DIR}/zstd_v04.h
${LIBRARY_LEGACY_DIR}/zstd_v05.h
${LIBRARY_LEGACY_DIR}/zstd_v06.h
${LIBRARY_LEGACY_DIR}/zstd_v07.h)
endif ()
if (MSVC)
set(MSVC_RESOURCE_DIR ${ZSTD_SOURCE_DIR}/build/VS2010/libzstd-dll)
set(PlatformDependResources ${MSVC_RESOURCE_DIR}/libzstd-dll.rc)
endif ()
# Explicitly set the language to C for all files, including ASM files.
# Our assembly expects to be compiled by a C compiler, and is only enabled for
# __GNUC__ compatible compilers. Otherwise all the ASM code is disabled by
# macros.
set_source_files_properties(${Sources} PROPERTIES LANGUAGE C)
# Split project to static and shared libraries build
set(library_targets)
if (ZSTD_BUILD_SHARED)
add_library(libzstd_shared SHARED ${Sources} ${Headers} ${PlatformDependResources})
list(APPEND library_targets libzstd_shared)
if (ZSTD_MULTITHREAD_SUPPORT)
set_property(TARGET libzstd_shared APPEND PROPERTY COMPILE_DEFINITIONS "ZSTD_MULTITHREAD")
if (UNIX)
target_link_libraries(libzstd_shared ${THREADS_LIBS})
endif ()
endif()
endif ()
if (ZSTD_BUILD_STATIC)
add_library(libzstd_static STATIC ${Sources} ${Headers})
list(APPEND library_targets libzstd_static)
if (ZSTD_MULTITHREAD_SUPPORT)
set_property(TARGET libzstd_static APPEND PROPERTY COMPILE_DEFINITIONS "ZSTD_MULTITHREAD")
if (UNIX)
target_link_libraries(libzstd_static ${THREADS_LIBS})
endif ()
endif ()
endif ()
# Add specific compile definitions for MSVC project
if (MSVC)
if (ZSTD_BUILD_SHARED)
set_property(TARGET libzstd_shared APPEND PROPERTY COMPILE_DEFINITIONS "ZSTD_DLL_EXPORT=1;ZSTD_HEAPMODE=0;_CONSOLE;_CRT_SECURE_NO_WARNINGS")
endif ()
if (ZSTD_BUILD_STATIC)
set_property(TARGET libzstd_static APPEND PROPERTY COMPILE_DEFINITIONS "ZSTD_HEAPMODE=0;_CRT_SECURE_NO_WARNINGS")
endif ()
endif ()
# With MSVC static library needs to be renamed to avoid conflict with import library
if (MSVC OR (WIN32 AND CMAKE_CXX_COMPILER_ID STREQUAL "Clang" AND NOT MINGW))
set(STATIC_LIBRARY_BASE_NAME zstd_static)
else ()
set(STATIC_LIBRARY_BASE_NAME zstd)
endif ()
# Define static and shared library names
if (ZSTD_BUILD_SHARED)
set_target_properties(
libzstd_shared
PROPERTIES
OUTPUT_NAME zstd
VERSION ${zstd_VERSION_MAJOR}.${zstd_VERSION_MINOR}.${zstd_VERSION_PATCH}
SOVERSION ${zstd_VERSION_MAJOR})
endif ()
if (ZSTD_BUILD_STATIC)
set_target_properties(
libzstd_static
PROPERTIES
POSITION_INDEPENDENT_CODE On
OUTPUT_NAME ${STATIC_LIBRARY_BASE_NAME})
endif ()
# pkg-config
include(JoinPaths) # can be replaced by cmake_path(APPEND) in CMake 3.20
set(PREFIX "${CMAKE_INSTALL_PREFIX}")
set(EXEC_PREFIX "\${prefix}")
join_paths(LIBDIR "\${exec_prefix}" "${CMAKE_INSTALL_LIBDIR}")
join_paths(INCLUDEDIR "\${prefix}" "${CMAKE_INSTALL_INCLUDEDIR}")
set(LIBS_PRIVATE "${THREADS_LIBS}")
set(VERSION "${zstd_VERSION}")
configure_file("${LIBRARY_DIR}/libzstd.pc.in" "${CMAKE_CURRENT_BINARY_DIR}/libzstd.pc" @ONLY)
install(FILES "${CMAKE_CURRENT_BINARY_DIR}/libzstd.pc" DESTINATION "${CMAKE_INSTALL_LIBDIR}/pkgconfig")
# install target
install(FILES
"${LIBRARY_DIR}/zstd.h"
"${LIBRARY_DIR}/zdict.h"
"${LIBRARY_DIR}/zstd_errors.h"
DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}")
install(TARGETS ${library_targets}
EXPORT zstdExports
INCLUDES DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}"
ARCHIVE DESTINATION "${CMAKE_INSTALL_LIBDIR}"
LIBRARY DESTINATION "${CMAKE_INSTALL_LIBDIR}"
RUNTIME DESTINATION "${CMAKE_INSTALL_BINDIR}"
BUNDLE DESTINATION "${CMAKE_INSTALL_BINDIR}"
)
# uninstall target
if (NOT TARGET uninstall)
configure_file(
"${CMAKE_CURRENT_SOURCE_DIR}/cmake_uninstall.cmake.in"
"${CMAKE_CURRENT_BINARY_DIR}/cmake_uninstall.cmake"
IMMEDIATE @ONLY)
add_custom_target(uninstall
COMMAND ${CMAKE_COMMAND} -P ${CMAKE_CURRENT_BINARY_DIR}/cmake_uninstall.cmake)
endif ()

22
raytracer/nvpro_core/third_party/zstd/build/cmake/lib/cmake_uninstall.cmake.in vendored Normal file
View file

@ -0,0 +1,22 @@
if(NOT EXISTS "@CMAKE_BINARY_DIR@/install_manifest.txt")
message(FATAL_ERROR "Cannot find install manifest: @CMAKE_BINARY_DIR@/install_manifest.txt")
endif()
file(READ "@CMAKE_BINARY_DIR@/install_manifest.txt" files)
string(REGEX REPLACE "\n" ";" files "${files}")
foreach(file ${files})
message(STATUS "Uninstalling $ENV{DESTDIR}${file}")
if(IS_SYMLINK "$ENV{DESTDIR}${file}" OR EXISTS "$ENV{DESTDIR}${file}")
exec_program(
"@CMAKE_COMMAND@" ARGS "-E remove \"$ENV{DESTDIR}${file}\""
OUTPUT_VARIABLE rm_out
RETURN_VALUE rm_retval
)
if(NOT "${rm_retval}" STREQUAL 0)
message(FATAL_ERROR "Problem when removing $ENV{DESTDIR}${file}")
endif()
else()
message(STATUS "File $ENV{DESTDIR}${file} does not exist.")
endif()
endforeach()

137
raytracer/nvpro_core/third_party/zstd/build/cmake/programs/CMakeLists.txt vendored Normal file
View file

@ -0,0 +1,137 @@
# ################################################################
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under both the BSD-style license (found in the
# LICENSE file in the root directory of this source tree) and the GPLv2 (found
# in the COPYING file in the root directory of this source tree).
# ################################################################
project(programs C)
set(CMAKE_INCLUDE_CURRENT_DIR TRUE)
# Define programs directory, where sources and header files are located
set(LIBRARY_DIR ${ZSTD_SOURCE_DIR}/lib)
set(PROGRAMS_DIR ${ZSTD_SOURCE_DIR}/programs)
include_directories(${PROGRAMS_DIR} ${LIBRARY_DIR} ${LIBRARY_DIR}/common ${LIBRARY_DIR}/compress ${LIBRARY_DIR}/dictBuilder)
if (ZSTD_LEGACY_SUPPORT)
set(PROGRAMS_LEGACY_DIR ${PROGRAMS_DIR}/legacy)
include_directories(${PROGRAMS_LEGACY_DIR} ${LIBRARY_DIR}/legacy)
endif ()
if (ZSTD_PROGRAMS_LINK_SHARED)
set(PROGRAMS_ZSTD_LINK_TARGET libzstd_shared)
else ()
set(PROGRAMS_ZSTD_LINK_TARGET libzstd_static)
endif ()
if (MSVC)
set(MSVC_RESOURCE_DIR ${ZSTD_SOURCE_DIR}/build/VS2010/zstd)
set(PlatformDependResources ${MSVC_RESOURCE_DIR}/zstd.rc)
endif ()
add_executable(zstd ${PROGRAMS_DIR}/zstdcli.c ${PROGRAMS_DIR}/util.c ${PROGRAMS_DIR}/timefn.c ${PROGRAMS_DIR}/fileio.c ${PROGRAMS_DIR}/fileio_asyncio.c ${PROGRAMS_DIR}/benchfn.c ${PROGRAMS_DIR}/benchzstd.c ${PROGRAMS_DIR}/datagen.c ${PROGRAMS_DIR}/dibio.c ${PROGRAMS_DIR}/zstdcli_trace.c ${PlatformDependResources})
target_link_libraries(zstd ${PROGRAMS_ZSTD_LINK_TARGET})
if (CMAKE_SYSTEM_NAME MATCHES "(Solaris|SunOS)")
target_link_libraries(zstd rt)
endif ()
install(TARGETS zstd
RUNTIME DESTINATION "${CMAKE_INSTALL_BINDIR}"
BUNDLE DESTINATION "${CMAKE_INSTALL_BINDIR}")
if (UNIX)
add_custom_target(zstdcat ALL ${CMAKE_COMMAND} -E create_symlink zstd zstdcat DEPENDS zstd COMMENT "Creating zstdcat symlink")
add_custom_target(unzstd ALL ${CMAKE_COMMAND} -E create_symlink zstd unzstd DEPENDS zstd COMMENT "Creating unzstd symlink")
install(FILES ${CMAKE_CURRENT_BINARY_DIR}/zstdcat DESTINATION "${CMAKE_INSTALL_BINDIR}")
install(FILES ${CMAKE_CURRENT_BINARY_DIR}/unzstd DESTINATION "${CMAKE_INSTALL_BINDIR}")
install(PROGRAMS ${PROGRAMS_DIR}/zstdgrep DESTINATION "${CMAKE_INSTALL_BINDIR}")
install(PROGRAMS ${PROGRAMS_DIR}/zstdless DESTINATION "${CMAKE_INSTALL_BINDIR}")
add_custom_target(zstd.1 ALL
${CMAKE_COMMAND} -E copy ${PROGRAMS_DIR}/zstd.1 .
COMMENT "Copying manpage zstd.1")
add_custom_target(zstdgrep.1 ALL
${CMAKE_COMMAND} -E copy ${PROGRAMS_DIR}/zstdgrep.1 .
COMMENT "Copying manpage zstdgrep.1")
add_custom_target(zstdless.1 ALL
${CMAKE_COMMAND} -E copy ${PROGRAMS_DIR}/zstdless.1 .
COMMENT "Copying manpage zstdless.1")
add_custom_target(zstdcat.1 ALL ${CMAKE_COMMAND} -E create_symlink zstd.1 zstdcat.1 DEPENDS zstd.1 COMMENT "Creating zstdcat.1 symlink")
add_custom_target(unzstd.1 ALL ${CMAKE_COMMAND} -E create_symlink zstd.1 unzstd.1 DEPENDS zstd.1 COMMENT "Creating unzstd.1 symlink")
# Define MAN_INSTALL_DIR if necessary
if (MAN_INSTALL_DIR)
else ()
set(MAN_INSTALL_DIR ${CMAKE_INSTALL_MANDIR}/man1)
endif ()
install(FILES
${CMAKE_CURRENT_BINARY_DIR}/zstd.1
${CMAKE_CURRENT_BINARY_DIR}/zstdcat.1
${CMAKE_CURRENT_BINARY_DIR}/unzstd.1
${CMAKE_CURRENT_BINARY_DIR}/zstdgrep.1
${CMAKE_CURRENT_BINARY_DIR}/zstdless.1
DESTINATION "${MAN_INSTALL_DIR}")
add_executable(zstd-frugal ${PROGRAMS_DIR}/zstdcli.c ${PROGRAMS_DIR}/util.c ${PROGRAMS_DIR}/timefn.c ${PROGRAMS_DIR}/fileio.c ${PROGRAMS_DIR}/fileio_asyncio.c)
target_link_libraries(zstd-frugal ${PROGRAMS_ZSTD_LINK_TARGET})
set_property(TARGET zstd-frugal APPEND PROPERTY COMPILE_DEFINITIONS "ZSTD_NOBENCH;ZSTD_NODICT;ZSTD_NOTRACE")
endif ()
# Add multi-threading support definitions
if (ZSTD_MULTITHREAD_SUPPORT)
set_property(TARGET zstd APPEND PROPERTY COMPILE_DEFINITIONS "ZSTD_MULTITHREAD")
if (UNIX)
target_link_libraries(zstd ${THREADS_LIBS})
add_custom_target(zstdmt ALL ${CMAKE_COMMAND} -E create_symlink zstd zstdmt DEPENDS zstd COMMENT "Creating zstdmt symlink")
install(FILES ${CMAKE_CURRENT_BINARY_DIR}/zstdmt DESTINATION "${CMAKE_INSTALL_BINDIR}")
endif ()
endif ()
option(ZSTD_ZLIB_SUPPORT "ZLIB SUPPORT" OFF)
option(ZSTD_LZMA_SUPPORT "LZMA SUPPORT" OFF)
option(ZSTD_LZ4_SUPPORT "LZ4 SUPPORT" OFF)
# Add gzip support
if (ZSTD_ZLIB_SUPPORT)
find_package(ZLIB REQUIRED)
if (ZLIB_FOUND)
include_directories(${ZLIB_INCLUDE_DIRS})
target_link_libraries(zstd ${ZLIB_LIBRARIES})
set_property(TARGET zstd APPEND PROPERTY COMPILE_DEFINITIONS "ZSTD_GZCOMPRESS;ZSTD_GZDECOMPRESS")
else ()
message(SEND_ERROR "zlib library is missing")
endif ()
endif ()
# Add lzma support
if (ZSTD_LZMA_SUPPORT)
find_package(LibLZMA REQUIRED)
if (LIBLZMA_FOUND)
include_directories(${LIBLZMA_INCLUDE_DIRS})
target_link_libraries(zstd ${LIBLZMA_LIBRARIES})
set_property(TARGET zstd APPEND PROPERTY COMPILE_DEFINITIONS "ZSTD_LZMACOMPRESS;ZSTD_LZMADECOMPRESS")
else ()
message(SEND_ERROR "lzma library is missing")
endif ()
endif ()
# Add lz4 support
if (ZSTD_LZ4_SUPPORT)
find_package(LibLZ4 REQUIRED)
if (LIBLZ4_FOUND)
include_directories(${LIBLZ4_INCLUDE_DIRS})
target_link_libraries(zstd ${LIBLZ4_LIBRARIES})
set_property(TARGET zstd APPEND PROPERTY COMPILE_DEFINITIONS "ZSTD_LZ4COMPRESS;ZSTD_LZ4DECOMPRESS")
else ()
message(SEND_ERROR "lz4 library is missing")
endif ()
endif ()

118
raytracer/nvpro_core/third_party/zstd/build/cmake/tests/CMakeLists.txt vendored Normal file
View file

@ -0,0 +1,118 @@
# ################################################################
# zstd - Makefile
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# BSD license
#
# Redistribution and use in source and binary forms, with or without modification,
# are permitted provided that the following conditions are met:
#
# * Redistributions of source code must retain the above copyright notice, this
# list of conditions and the following disclaimer.
#
# * Redistributions in binary form must reproduce the above copyright notice, this
# list of conditions and the following disclaimer in the documentation and/or
# other materials provided with the distribution.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR
# ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
# (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
# LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
# ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
# SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
# You can contact the author at :
# - zstd homepage : https://facebook.github.io/zstd/
# ################################################################
project(tests)
# name: Cache variable name. The value is expected to be a semicolon-separated
# list of command line flags
# default_value: Value to initialize the option with. Can be space separated.
function(AddTestFlagsOption name default_value doc)
string(STRIP "${default_value}" default_value)
string(REGEX REPLACE " +" ";" default_value "${default_value}")
set(${name} ${default_value} CACHE STRING "${doc}")
mark_as_advanced(${name})
endfunction()
set(CMAKE_INCLUDE_CURRENT_DIR TRUE)
# Define programs directory, where sources and header files are located
set(LIBRARY_DIR ${ZSTD_SOURCE_DIR}/lib)
set(PROGRAMS_DIR ${ZSTD_SOURCE_DIR}/programs)
set(TESTS_DIR ${ZSTD_SOURCE_DIR}/tests)
include_directories(${TESTS_DIR} ${PROGRAMS_DIR} ${LIBRARY_DIR} ${LIBRARY_DIR}/common ${LIBRARY_DIR}/compress ${LIBRARY_DIR}/dictBuilder)
add_executable(datagen ${PROGRAMS_DIR}/datagen.c ${TESTS_DIR}/datagencli.c)
target_link_libraries(datagen libzstd_static)
#
# fullbench
#
add_executable(fullbench ${PROGRAMS_DIR}/datagen.c ${PROGRAMS_DIR}/util.c ${PROGRAMS_DIR}/timefn.c ${PROGRAMS_DIR}/benchfn.c ${PROGRAMS_DIR}/benchzstd.c ${TESTS_DIR}/fullbench.c)
if (NOT MSVC)
target_compile_options(fullbench PRIVATE "-Wno-deprecated-declarations")
endif()
target_link_libraries(fullbench libzstd_static)
add_test(NAME fullbench COMMAND fullbench ${ZSTD_FULLBENCH_FLAGS})
#
# fuzzer
#
add_executable(fuzzer ${PROGRAMS_DIR}/datagen.c ${PROGRAMS_DIR}/util.c ${PROGRAMS_DIR}/timefn.c ${TESTS_DIR}/fuzzer.c)
if (NOT MSVC)
target_compile_options(fuzzer PRIVATE "-Wno-deprecated-declarations")
endif()
target_link_libraries(fuzzer libzstd_static)
AddTestFlagsOption(ZSTD_FUZZER_FLAGS "$ENV{FUZZERTEST} $ENV{FUZZER_FLAGS}"
"Semicolon-separated list of flags to pass to the fuzzer test (see `fuzzer -h` for usage)")
add_test(NAME fuzzer COMMAND fuzzer ${ZSTD_FUZZER_FLAGS})
# Disable the timeout since the run time is too long for the default timeout of
# 1500 seconds and varies considerably between low-end and high-end CPUs.
# set_tests_properties(fuzzer PROPERTIES TIMEOUT 0)
#
# zstreamtest
#
add_executable(zstreamtest ${PROGRAMS_DIR}/datagen.c ${PROGRAMS_DIR}/util.c ${PROGRAMS_DIR}/timefn.c ${TESTS_DIR}/seqgen.c ${TESTS_DIR}/zstreamtest.c ${TESTS_DIR}/external_matchfinder.c)
if (NOT MSVC)
target_compile_options(zstreamtest PRIVATE "-Wno-deprecated-declarations")
endif()
target_link_libraries(zstreamtest libzstd_static)
AddTestFlagsOption(ZSTD_ZSTREAM_FLAGS "$ENV{ZSTREAM_TESTTIME} $ENV{FUZZER_FLAGS}"
"Semicolon-separated list of flags to pass to the zstreamtest test (see `zstreamtest -h` for usage)")
add_test(NAME zstreamtest COMMAND zstreamtest ${ZSTD_ZSTREAM_FLAGS})
#
# playTests.sh
#
AddTestFlagsOption(ZSTD_PLAYTESTS_FLAGS "$ENV{PLAYTESTS_FLAGS}"
"Semicolon-separated list of flags to pass to the playTests.sh test")
add_test(NAME playTests COMMAND sh -c "\"${TESTS_DIR}/playTests.sh\" ${ZSTD_PLAYTESTS_FLAGS}")
find_program(UNAME uname) # Run script only in unix shell environments
if (ZSTD_BUILD_PROGRAMS AND UNAME)
set_property(TEST playTests APPEND PROPERTY ENVIRONMENT
"ZSTD_BIN=$<TARGET_FILE:zstd>"
"DATAGEN_BIN=$<TARGET_FILE:datagen>"
)
else()
message(STATUS "Disabling playTests.sh test because requirements not met")
set_tests_properties(playTests PROPERTIES DISABLED YES)
endif()
# Label the "Medium" set of tests (see TESTING.md)
set_property(TEST fuzzer zstreamtest playTests APPEND PROPERTY LABELS Medium)
add_executable(paramgrill ${PROGRAMS_DIR}/benchfn.c ${PROGRAMS_DIR}/benchzstd.c ${PROGRAMS_DIR}/datagen.c ${PROGRAMS_DIR}/util.c ${PROGRAMS_DIR}/timefn.c ${TESTS_DIR}/paramgrill.c)
if (UNIX)
target_link_libraries(paramgrill libzstd_static m) #m is math library
else()
target_link_libraries(paramgrill libzstd_static)
endif ()

1
raytracer/nvpro_core/third_party/zstd/build/cmake/zstdConfig.cmake vendored Normal file
View file

@ -0,0 +1 @@
include("${CMAKE_CURRENT_LIST_DIR}/zstdTargets.cmake")

224
raytracer/nvpro_core/third_party/zstd/lib/README.md vendored Normal file
View file

@ -0,0 +1,224 @@
Zstandard library files
================================
The __lib__ directory is split into several sub-directories,
in order to make it easier to select or exclude features.
#### Building
`Makefile` script is provided, supporting [Makefile conventions](https://www.gnu.org/prep/standards/html_node/Makefile-Conventions.html#Makefile-Conventions),
including commands variables, staged install, directory variables and standard targets.
- `make` : generates both static and dynamic libraries
- `make install` : install libraries and headers in target system directories
`libzstd` default scope is pretty large, including compression, decompression, dictionary builder,
and support for decoding legacy formats >= v0.5.0.
The scope can be reduced on demand (see paragraph _modular build_).
#### Multithreading support
When building with `make`, by default the dynamic library is multithreaded and static library is single-threaded (for compatibility reasons).
Enabling multithreading requires 2 conditions :
- set build macro `ZSTD_MULTITHREAD` (`-DZSTD_MULTITHREAD` for `gcc`)
- for POSIX systems : compile with pthread (`-pthread` compilation flag for `gcc`)
For convenience, we provide a build target to generate multi and single threaded libraries:
- Force enable multithreading on both dynamic and static libraries by appending `-mt` to the target, e.g. `make lib-mt`.
- Force disable multithreading on both dynamic and static libraries by appending `-nomt` to the target, e.g. `make lib-nomt`.
- By default, as mentioned before, dynamic library is multithreaded, and static library is single-threaded, e.g. `make lib`.
When linking a POSIX program with a multithreaded version of `libzstd`,
note that it's necessary to invoke the `-pthread` flag during link stage.
Multithreading capabilities are exposed
via the [advanced API defined in `lib/zstd.h`](https://github.com/facebook/zstd/blob/v1.4.3/lib/zstd.h#L351).
#### API
Zstandard's stable API is exposed within [lib/zstd.h](zstd.h).
#### Advanced API
Optional advanced features are exposed via :
- `lib/zstd_errors.h` : translates `size_t` function results
into a `ZSTD_ErrorCode`, for accurate error handling.
- `ZSTD_STATIC_LINKING_ONLY` : if this macro is defined _before_ including `zstd.h`,
it unlocks access to the experimental API,
exposed in the second part of `zstd.h`.
All definitions in the experimental APIs are unstable,
they may still change in the future, or even be removed.
As a consequence, experimental definitions shall ___never be used with dynamic library___ !
Only static linking is allowed.
#### Modular build
It's possible to compile only a limited set of features within `libzstd`.
The file structure is designed to make this selection manually achievable for any build system :
- Directory `lib/common` is always required, for all variants.
- Compression source code lies in `lib/compress`
- Decompression source code lies in `lib/decompress`
- It's possible to include only `compress` or only `decompress`, they don't depend on each other.
- `lib/dictBuilder` : makes it possible to generate dictionaries from a set of samples.
The API is exposed in `lib/dictBuilder/zdict.h`.
This module depends on both `lib/common` and `lib/compress` .
- `lib/legacy` : makes it possible to decompress legacy zstd formats, starting from `v0.1.0`.
This module depends on `lib/common` and `lib/decompress`.
To enable this feature, define `ZSTD_LEGACY_SUPPORT` during compilation.
Specifying a number limits versions supported to that version onward.
For example, `ZSTD_LEGACY_SUPPORT=2` means : "support legacy formats >= v0.2.0".
Conversely, `ZSTD_LEGACY_SUPPORT=0` means "do __not__ support legacy formats".
By default, this build macro is set as `ZSTD_LEGACY_SUPPORT=5`.
Decoding supported legacy format is a transparent capability triggered within decompression functions.
It's also allowed to invoke legacy API directly, exposed in `lib/legacy/zstd_legacy.h`.
Each version does also provide its own set of advanced API.
For example, advanced API for version `v0.4` is exposed in `lib/legacy/zstd_v04.h` .
- While invoking `make libzstd`, it's possible to define build macros
`ZSTD_LIB_COMPRESSION, ZSTD_LIB_DECOMPRESSION`, `ZSTD_LIB_DICTBUILDER`,
and `ZSTD_LIB_DEPRECATED` as `0` to forgo compilation of the
corresponding features. This will also disable compilation of all
dependencies (e.g. `ZSTD_LIB_COMPRESSION=0` will also disable
dictBuilder).
- There are a number of options that can help minimize the binary size of
`libzstd`.
The first step is to select the components needed (using the above-described
`ZSTD_LIB_COMPRESSION` etc.).
The next step is to set `ZSTD_LIB_MINIFY` to `1` when invoking `make`. This
disables various optional components and changes the compilation flags to
prioritize space-saving.
Detailed options: Zstandard's code and build environment is set up by default
to optimize above all else for performance. In pursuit of this goal, Zstandard
makes significant trade-offs in code size. For example, Zstandard often has
more than one implementation of a particular component, with each
implementation optimized for different scenarios. For example, the Huffman
decoder has complementary implementations that decode the stream one symbol at
a time or two symbols at a time. Zstd normally includes both (and dispatches
between them at runtime), but by defining `HUF_FORCE_DECOMPRESS_X1` or
`HUF_FORCE_DECOMPRESS_X2`, you can force the use of one or the other, avoiding
compilation of the other. Similarly, `ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT`
and `ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG` force the compilation and use of
only one or the other of two decompression implementations. The smallest
binary is achieved by using `HUF_FORCE_DECOMPRESS_X1` and
`ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT` (implied by `ZSTD_LIB_MINIFY`).
For squeezing the last ounce of size out, you can also define
`ZSTD_NO_INLINE`, which disables inlining, and `ZSTD_STRIP_ERROR_STRINGS`,
which removes the error messages that are otherwise returned by
`ZSTD_getErrorName` (implied by `ZSTD_LIB_MINIFY`).
Finally, when integrating into your application, make sure you're doing link-
time optimization and unused symbol garbage collection (via some combination of,
e.g., `-flto`, `-ffat-lto-objects`, `-fuse-linker-plugin`,
`-ffunction-sections`, `-fdata-sections`, `-fmerge-all-constants`,
`-Wl,--gc-sections`, `-Wl,-z,norelro`, and an archiver that understands
the compiler's intermediate representation, e.g., `AR=gcc-ar`). Consult your
compiler's documentation.
- While invoking `make libzstd`, the build macro `ZSTD_LEGACY_MULTITHREADED_API=1`
will expose the deprecated `ZSTDMT` API exposed by `zstdmt_compress.h` in
the shared library, which is now hidden by default.
- The build macro `DYNAMIC_BMI2` can be set to 1 or 0 in order to generate binaries
which can detect at runtime the presence of BMI2 instructions, and use them only if present.
These instructions contribute to better performance, notably on the decoder side.
By default, this feature is automatically enabled on detecting
the right instruction set (x64) and compiler (clang or gcc >= 5).
It's obviously disabled for different cpus,
or when BMI2 instruction set is _required_ by the compiler command line
(in this case, only the BMI2 code path is generated).
Setting this macro will either force to generate the BMI2 dispatcher (1)
or prevent it (0). It overrides automatic detection.
- The build macro `ZSTD_NO_UNUSED_FUNCTIONS` can be defined to hide the definitions of functions
that zstd does not use. Not all unused functions are hidden, but they can be if needed.
Currently, this macro will hide function definitions in FSE and HUF that use an excessive
amount of stack space.
- The build macro `ZSTD_NO_INTRINSICS` can be defined to disable all explicit intrinsics.
Compiler builtins are still used.
- The build macro `ZSTD_DECODER_INTERNAL_BUFFER` can be set to control
the amount of extra memory used during decompression to store literals.
This defaults to 64kB. Reducing this value reduces the memory footprint of
`ZSTD_DCtx` decompression contexts,
but might also result in a small decompression speed cost.
- The C compiler macros `ZSTDLIB_VISIBLE`, `ZSTDERRORLIB_VISIBLE` and `ZDICTLIB_VISIBLE`
can be overridden to control the visibility of zstd's API. Additionally,
`ZSTDLIB_STATIC_API` and `ZDICTLIB_STATIC_API` can be overridden to control the visibility
of zstd's static API. Specifically, it can be set to `ZSTDLIB_HIDDEN` to hide the symbols
from the shared library. These macros default to `ZSTDLIB_VISIBILITY`,
`ZSTDERRORLIB_VSIBILITY`, and `ZDICTLIB_VISIBILITY` if unset, for backwards compatibility
with the old macro names.
#### Windows : using MinGW+MSYS to create DLL
DLL can be created using MinGW+MSYS with the `make libzstd` command.
This command creates `dll\libzstd.dll` and the import library `dll\libzstd.lib`.
The import library is only required with Visual C++.
The header file `zstd.h` and the dynamic library `dll\libzstd.dll` are required to
compile a project using gcc/MinGW.
The dynamic library has to be added to linking options.
It means that if a project that uses ZSTD consists of a single `test-dll.c`
file it should be linked with `dll\libzstd.dll`. For example:
```
gcc $(CFLAGS) -Iinclude/ test-dll.c -o test-dll dll\libzstd.dll
```
The compiled executable will require ZSTD DLL which is available at `dll\libzstd.dll`.
#### Advanced Build options
The build system requires a hash function in order to
separate object files created with different compilation flags.
By default, it tries to use `md5sum` or equivalent.
The hash function can be manually switched by setting the `HASH` variable.
For example : `make HASH=xxhsum`
The hash function needs to generate at least 64-bit using hexadecimal format.
When no hash function is found,
the Makefile just generates all object files into the same default directory,
irrespective of compilation flags.
This functionality only matters if `libzstd` is compiled multiple times
with different build flags.
The build directory, where object files are stored
can also be manually controlled using variable `BUILD_DIR`,
for example `make BUILD_DIR=objectDir/v1`.
In which case, the hash function doesn't matter.
#### Deprecated API
Obsolete API on their way out are stored in directory `lib/deprecated`.
At this stage, it contains older streaming prototypes, in `lib/deprecated/zbuff.h`.
These prototypes will be removed in some future version.
Consider migrating code towards supported streaming API exposed in `zstd.h`.
#### Miscellaneous
The other files are not source code. There are :
- `BUCK` : support for `buck` build system (https://buckbuild.com/)
- `Makefile` : `make` script to build and install zstd library (static and dynamic)
- `README.md` : this file
- `dll/` : resources directory for Windows compilation
- `libzstd.pc.in` : script for `pkg-config` (used in `make install`)

55
raytracer/nvpro_core/third_party/zstd/lib/common/allocations.h vendored Normal file
View file

@ -0,0 +1,55 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* This file provides custom allocation primitives
*/
#define ZSTD_DEPS_NEED_MALLOC
#include "zstd_deps.h" /* ZSTD_malloc, ZSTD_calloc, ZSTD_free, ZSTD_memset */
#include "mem.h" /* MEM_STATIC */
#define ZSTD_STATIC_LINKING_ONLY
#include "../zstd.h" /* ZSTD_customMem */
#ifndef ZSTD_ALLOCATIONS_H
#define ZSTD_ALLOCATIONS_H
/* custom memory allocation functions */
MEM_STATIC void* ZSTD_customMalloc(size_t size, ZSTD_customMem customMem)
{
if (customMem.customAlloc)
return customMem.customAlloc(customMem.opaque, size);
return ZSTD_malloc(size);
}
MEM_STATIC void* ZSTD_customCalloc(size_t size, ZSTD_customMem customMem)
{
if (customMem.customAlloc) {
/* calloc implemented as malloc+memset;
* not as efficient as calloc, but next best guess for custom malloc */
void* const ptr = customMem.customAlloc(customMem.opaque, size);
ZSTD_memset(ptr, 0, size);
return ptr;
}
return ZSTD_calloc(1, size);
}
MEM_STATIC void ZSTD_customFree(void* ptr, ZSTD_customMem customMem)
{
if (ptr!=NULL) {
if (customMem.customFree)
customMem.customFree(customMem.opaque, ptr);
else
ZSTD_free(ptr);
}
}
#endif /* ZSTD_ALLOCATIONS_H */

200
raytracer/nvpro_core/third_party/zstd/lib/common/bits.h vendored Normal file
View file

@ -0,0 +1,200 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_BITS_H
#define ZSTD_BITS_H
#include "mem.h"
MEM_STATIC unsigned ZSTD_countTrailingZeros32_fallback(U32 val)
{
assert(val != 0);
{
static const U32 DeBruijnBytePos[32] = {0, 1, 28, 2, 29, 14, 24, 3,
30, 22, 20, 15, 25, 17, 4, 8,
31, 27, 13, 23, 21, 19, 16, 7,
26, 12, 18, 6, 11, 5, 10, 9};
return DeBruijnBytePos[((U32) ((val & -(S32) val) * 0x077CB531U)) >> 27];
}
}
MEM_STATIC unsigned ZSTD_countTrailingZeros32(U32 val)
{
assert(val != 0);
# if defined(_MSC_VER)
# if STATIC_BMI2 == 1
return (unsigned)_tzcnt_u32(val);
# else
if (val != 0) {
unsigned long r;
_BitScanForward(&r, val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return (unsigned)__builtin_ctz(val);
# else
return ZSTD_countTrailingZeros32_fallback(val);
# endif
}
MEM_STATIC unsigned ZSTD_countLeadingZeros32_fallback(U32 val) {
assert(val != 0);
{
static const U32 DeBruijnClz[32] = {0, 9, 1, 10, 13, 21, 2, 29,
11, 14, 16, 18, 22, 25, 3, 30,
8, 12, 20, 28, 15, 17, 24, 7,
19, 27, 23, 6, 26, 5, 4, 31};
val |= val >> 1;
val |= val >> 2;
val |= val >> 4;
val |= val >> 8;
val |= val >> 16;
return 31 - DeBruijnClz[(val * 0x07C4ACDDU) >> 27];
}
}
MEM_STATIC unsigned ZSTD_countLeadingZeros32(U32 val)
{
assert(val != 0);
# if defined(_MSC_VER)
# if STATIC_BMI2 == 1
return (unsigned)_lzcnt_u32(val);
# else
if (val != 0) {
unsigned long r;
_BitScanReverse(&r, val);
return (unsigned)(31 - r);
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return (unsigned)__builtin_clz(val);
# else
return ZSTD_countLeadingZeros32_fallback(val);
# endif
}
MEM_STATIC unsigned ZSTD_countTrailingZeros64(U64 val)
{
assert(val != 0);
# if defined(_MSC_VER) && defined(_WIN64)
# if STATIC_BMI2 == 1
return (unsigned)_tzcnt_u64(val);
# else
if (val != 0) {
unsigned long r;
_BitScanForward64(&r, val);
return (unsigned)r;
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4) && defined(__LP64__)
return (unsigned)__builtin_ctzll(val);
# else
{
U32 mostSignificantWord = (U32)(val >> 32);
U32 leastSignificantWord = (U32)val;
if (leastSignificantWord == 0) {
return 32 + ZSTD_countTrailingZeros32(mostSignificantWord);
} else {
return ZSTD_countTrailingZeros32(leastSignificantWord);
}
}
# endif
}
MEM_STATIC unsigned ZSTD_countLeadingZeros64(U64 val)
{
assert(val != 0);
# if defined(_MSC_VER) && defined(_WIN64)
# if STATIC_BMI2 == 1
return (unsigned)_lzcnt_u64(val);
# else
if (val != 0) {
unsigned long r;
_BitScanReverse64(&r, val);
return (unsigned)(63 - r);
} else {
/* Should not reach this code path */
__assume(0);
}
# endif
# elif defined(__GNUC__) && (__GNUC__ >= 4)
return (unsigned)(__builtin_clzll(val));
# else
{
U32 mostSignificantWord = (U32)(val >> 32);
U32 leastSignificantWord = (U32)val;
if (mostSignificantWord == 0) {
return 32 + ZSTD_countLeadingZeros32(leastSignificantWord);
} else {
return ZSTD_countLeadingZeros32(mostSignificantWord);
}
}
# endif
}
MEM_STATIC unsigned ZSTD_NbCommonBytes(size_t val)
{
if (MEM_isLittleEndian()) {
if (MEM_64bits()) {
return ZSTD_countTrailingZeros64((U64)val) >> 3;
} else {
return ZSTD_countTrailingZeros32((U32)val) >> 3;
}
} else { /* Big Endian CPU */
if (MEM_64bits()) {
return ZSTD_countLeadingZeros64((U64)val) >> 3;
} else {
return ZSTD_countLeadingZeros32((U32)val) >> 3;
}
}
}
MEM_STATIC unsigned ZSTD_highbit32(U32 val) /* compress, dictBuilder, decodeCorpus */
{
assert(val != 0);
return 31 - ZSTD_countLeadingZeros32(val);
}
/* ZSTD_rotateRight_*():
* Rotates a bitfield to the right by "count" bits.
* https://en.wikipedia.org/w/index.php?title=Circular_shift&oldid=991635599#Implementing_circular_shifts
*/
MEM_STATIC
U64 ZSTD_rotateRight_U64(U64 const value, U32 count) {
assert(count < 64);
count &= 0x3F; /* for fickle pattern recognition */
return (value >> count) | (U64)(value << ((0U - count) & 0x3F));
}
MEM_STATIC
U32 ZSTD_rotateRight_U32(U32 const value, U32 count) {
assert(count < 32);
count &= 0x1F; /* for fickle pattern recognition */
return (value >> count) | (U32)(value << ((0U - count) & 0x1F));
}
MEM_STATIC
U16 ZSTD_rotateRight_U16(U16 const value, U32 count) {
assert(count < 16);
count &= 0x0F; /* for fickle pattern recognition */
return (value >> count) | (U16)(value << ((0U - count) & 0x0F));
}
#endif /* ZSTD_BITS_H */

437
raytracer/nvpro_core/third_party/zstd/lib/common/bitstream.h vendored Normal file
View file

@ -0,0 +1,437 @@
/* ******************************************************************
* bitstream
* Part of FSE library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
#ifndef BITSTREAM_H_MODULE
#define BITSTREAM_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/*
* This API consists of small unitary functions, which must be inlined for best performance.
* Since link-time-optimization is not available for all compilers,
* these functions are defined into a .h to be included.
*/
/*-****************************************
* Dependencies
******************************************/
#include "mem.h" /* unaligned access routines */
#include "compiler.h" /* UNLIKELY() */
#include "debug.h" /* assert(), DEBUGLOG(), RAWLOG() */
#include "error_private.h" /* error codes and messages */
#include "bits.h" /* ZSTD_highbit32 */
/*=========================================
* Target specific
=========================================*/
#ifndef ZSTD_NO_INTRINSICS
# if (defined(__BMI__) || defined(__BMI2__)) && defined(__GNUC__)
# include <immintrin.h> /* support for bextr (experimental)/bzhi */
# elif defined(__ICCARM__)
# include <intrinsics.h>
# endif
#endif
#define STREAM_ACCUMULATOR_MIN_32 25
#define STREAM_ACCUMULATOR_MIN_64 57
#define STREAM_ACCUMULATOR_MIN ((U32)(MEM_32bits() ? STREAM_ACCUMULATOR_MIN_32 : STREAM_ACCUMULATOR_MIN_64))
/*-******************************************
* bitStream encoding API (write forward)
********************************************/
/* bitStream can mix input from multiple sources.
* A critical property of these streams is that they encode and decode in **reverse** direction.
* So the first bit sequence you add will be the last to be read, like a LIFO stack.
*/
typedef struct {
size_t bitContainer;
unsigned bitPos;
char* startPtr;
char* ptr;
char* endPtr;
} BIT_CStream_t;
MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC, void* dstBuffer, size_t dstCapacity);
MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC, size_t value, unsigned nbBits);
MEM_STATIC void BIT_flushBits(BIT_CStream_t* bitC);
MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC);
/* Start with initCStream, providing the size of buffer to write into.
* bitStream will never write outside of this buffer.
* `dstCapacity` must be >= sizeof(bitD->bitContainer), otherwise @return will be an error code.
*
* bits are first added to a local register.
* Local register is size_t, hence 64-bits on 64-bits systems, or 32-bits on 32-bits systems.
* Writing data into memory is an explicit operation, performed by the flushBits function.
* Hence keep track how many bits are potentially stored into local register to avoid register overflow.
* After a flushBits, a maximum of 7 bits might still be stored into local register.
*
* Avoid storing elements of more than 24 bits if you want compatibility with 32-bits bitstream readers.
*
* Last operation is to close the bitStream.
* The function returns the final size of CStream in bytes.
* If data couldn't fit into `dstBuffer`, it will return a 0 ( == not storable)
*/
/*-********************************************
* bitStream decoding API (read backward)
**********************************************/
typedef struct {
size_t bitContainer;
unsigned bitsConsumed;
const char* ptr;
const char* start;
const char* limitPtr;
} BIT_DStream_t;
typedef enum { BIT_DStream_unfinished = 0,
BIT_DStream_endOfBuffer = 1,
BIT_DStream_completed = 2,
BIT_DStream_overflow = 3 } BIT_DStream_status; /* result of BIT_reloadDStream() */
/* 1,2,4,8 would be better for bitmap combinations, but slows down performance a bit ... :( */
MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize);
MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits);
MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD);
MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* bitD);
/* Start by invoking BIT_initDStream().
* A chunk of the bitStream is then stored into a local register.
* Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
* You can then retrieve bitFields stored into the local register, **in reverse order**.
* Local register is explicitly reloaded from memory by the BIT_reloadDStream() method.
* A reload guarantee a minimum of ((8*sizeof(bitD->bitContainer))-7) bits when its result is BIT_DStream_unfinished.
* Otherwise, it can be less than that, so proceed accordingly.
* Checking if DStream has reached its end can be performed with BIT_endOfDStream().
*/
/*-****************************************
* unsafe API
******************************************/
MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC, size_t value, unsigned nbBits);
/* faster, but works only if value is "clean", meaning all high bits above nbBits are 0 */
MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC);
/* unsafe version; does not check buffer overflow */
MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits);
/* faster, but works only if nbBits >= 1 */
/*===== Local Constants =====*/
static const unsigned BIT_mask[] = {
0, 1, 3, 7, 0xF, 0x1F,
0x3F, 0x7F, 0xFF, 0x1FF, 0x3FF, 0x7FF,
0xFFF, 0x1FFF, 0x3FFF, 0x7FFF, 0xFFFF, 0x1FFFF,
0x3FFFF, 0x7FFFF, 0xFFFFF, 0x1FFFFF, 0x3FFFFF, 0x7FFFFF,
0xFFFFFF, 0x1FFFFFF, 0x3FFFFFF, 0x7FFFFFF, 0xFFFFFFF, 0x1FFFFFFF,
0x3FFFFFFF, 0x7FFFFFFF}; /* up to 31 bits */
#define BIT_MASK_SIZE (sizeof(BIT_mask) / sizeof(BIT_mask[0]))
/*-**************************************************************
* bitStream encoding
****************************************************************/
/*! BIT_initCStream() :
* `dstCapacity` must be > sizeof(size_t)
* @return : 0 if success,
* otherwise an error code (can be tested using ERR_isError()) */
MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC,
void* startPtr, size_t dstCapacity)
{
bitC->bitContainer = 0;
bitC->bitPos = 0;
bitC->startPtr = (char*)startPtr;
bitC->ptr = bitC->startPtr;
bitC->endPtr = bitC->startPtr + dstCapacity - sizeof(bitC->bitContainer);
if (dstCapacity <= sizeof(bitC->bitContainer)) return ERROR(dstSize_tooSmall);
return 0;
}
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits)
{
#if defined(STATIC_BMI2) && STATIC_BMI2 == 1 && !defined(ZSTD_NO_INTRINSICS)
return _bzhi_u64(bitContainer, nbBits);
#else
assert(nbBits < BIT_MASK_SIZE);
return bitContainer & BIT_mask[nbBits];
#endif
}
/*! BIT_addBits() :
* can add up to 31 bits into `bitC`.
* Note : does not check for register overflow ! */
MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC,
size_t value, unsigned nbBits)
{
DEBUG_STATIC_ASSERT(BIT_MASK_SIZE == 32);
assert(nbBits < BIT_MASK_SIZE);
assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8);
bitC->bitContainer |= BIT_getLowerBits(value, nbBits) << bitC->bitPos;
bitC->bitPos += nbBits;
}
/*! BIT_addBitsFast() :
* works only if `value` is _clean_,
* meaning all high bits above nbBits are 0 */
MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC,
size_t value, unsigned nbBits)
{
assert((value>>nbBits) == 0);
assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8);
bitC->bitContainer |= value << bitC->bitPos;
bitC->bitPos += nbBits;
}
/*! BIT_flushBitsFast() :
* assumption : bitContainer has not overflowed
* unsafe version; does not check buffer overflow */
MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC)
{
size_t const nbBytes = bitC->bitPos >> 3;
assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8);
assert(bitC->ptr <= bitC->endPtr);
MEM_writeLEST(bitC->ptr, bitC->bitContainer);
bitC->ptr += nbBytes;
bitC->bitPos &= 7;
bitC->bitContainer >>= nbBytes*8;
}
/*! BIT_flushBits() :
* assumption : bitContainer has not overflowed
* safe version; check for buffer overflow, and prevents it.
* note : does not signal buffer overflow.
* overflow will be revealed later on using BIT_closeCStream() */
MEM_STATIC void BIT_flushBits(BIT_CStream_t* bitC)
{
size_t const nbBytes = bitC->bitPos >> 3;
assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8);
assert(bitC->ptr <= bitC->endPtr);
MEM_writeLEST(bitC->ptr, bitC->bitContainer);
bitC->ptr += nbBytes;
if (bitC->ptr > bitC->endPtr) bitC->ptr = bitC->endPtr;
bitC->bitPos &= 7;
bitC->bitContainer >>= nbBytes*8;
}
/*! BIT_closeCStream() :
* @return : size of CStream, in bytes,
* or 0 if it could not fit into dstBuffer */
MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC)
{
BIT_addBitsFast(bitC, 1, 1); /* endMark */
BIT_flushBits(bitC);
if (bitC->ptr >= bitC->endPtr) return 0; /* overflow detected */
return (bitC->ptr - bitC->startPtr) + (bitC->bitPos > 0);
}
/*-********************************************************
* bitStream decoding
**********************************************************/
/*! BIT_initDStream() :
* Initialize a BIT_DStream_t.
* `bitD` : a pointer to an already allocated BIT_DStream_t structure.
* `srcSize` must be the *exact* size of the bitStream, in bytes.
* @return : size of stream (== srcSize), or an errorCode if a problem is detected
*/
MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize)
{
if (srcSize < 1) { ZSTD_memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); }
bitD->start = (const char*)srcBuffer;
bitD->limitPtr = bitD->start + sizeof(bitD->bitContainer);
if (srcSize >= sizeof(bitD->bitContainer)) { /* normal case */
bitD->ptr = (const char*)srcBuffer + srcSize - sizeof(bitD->bitContainer);
bitD->bitContainer = MEM_readLEST(bitD->ptr);
{ BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0; /* ensures bitsConsumed is always set */
if (lastByte == 0) return ERROR(GENERIC); /* endMark not present */ }
} else {
bitD->ptr = bitD->start;
bitD->bitContainer = *(const BYTE*)(bitD->start);
switch(srcSize)
{
case 7: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[6]) << (sizeof(bitD->bitContainer)*8 - 16);
ZSTD_FALLTHROUGH;
case 6: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[5]) << (sizeof(bitD->bitContainer)*8 - 24);
ZSTD_FALLTHROUGH;
case 5: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[4]) << (sizeof(bitD->bitContainer)*8 - 32);
ZSTD_FALLTHROUGH;
case 4: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[3]) << 24;
ZSTD_FALLTHROUGH;
case 3: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[2]) << 16;
ZSTD_FALLTHROUGH;
case 2: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[1]) << 8;
ZSTD_FALLTHROUGH;
default: break;
}
{ BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1];
bitD->bitsConsumed = lastByte ? 8 - ZSTD_highbit32(lastByte) : 0;
if (lastByte == 0) return ERROR(corruption_detected); /* endMark not present */
}
bitD->bitsConsumed += (U32)(sizeof(bitD->bitContainer) - srcSize)*8;
}
return srcSize;
}
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getUpperBits(size_t bitContainer, U32 const start)
{
return bitContainer >> start;
}
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits)
{
U32 const regMask = sizeof(bitContainer)*8 - 1;
/* if start > regMask, bitstream is corrupted, and result is undefined */
assert(nbBits < BIT_MASK_SIZE);
/* x86 transform & ((1 << nbBits) - 1) to bzhi instruction, it is better
* than accessing memory. When bmi2 instruction is not present, we consider
* such cpus old (pre-Haswell, 2013) and their performance is not of that
* importance.
*/
#if defined(__x86_64__) || defined(_M_X86)
return (bitContainer >> (start & regMask)) & ((((U64)1) << nbBits) - 1);
#else
return (bitContainer >> (start & regMask)) & BIT_mask[nbBits];
#endif
}
/*! BIT_lookBits() :
* Provides next n bits from local register.
* local register is not modified.
* On 32-bits, maxNbBits==24.
* On 64-bits, maxNbBits==56.
* @return : value extracted */
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_lookBits(const BIT_DStream_t* bitD, U32 nbBits)
{
/* arbitrate between double-shift and shift+mask */
#if 1
/* if bitD->bitsConsumed + nbBits > sizeof(bitD->bitContainer)*8,
* bitstream is likely corrupted, and result is undefined */
return BIT_getMiddleBits(bitD->bitContainer, (sizeof(bitD->bitContainer)*8) - bitD->bitsConsumed - nbBits, nbBits);
#else
/* this code path is slower on my os-x laptop */
U32 const regMask = sizeof(bitD->bitContainer)*8 - 1;
return ((bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> 1) >> ((regMask-nbBits) & regMask);
#endif
}
/*! BIT_lookBitsFast() :
* unsafe version; only works if nbBits >= 1 */
MEM_STATIC size_t BIT_lookBitsFast(const BIT_DStream_t* bitD, U32 nbBits)
{
U32 const regMask = sizeof(bitD->bitContainer)*8 - 1;
assert(nbBits >= 1);
return (bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> (((regMask+1)-nbBits) & regMask);
}
MEM_STATIC FORCE_INLINE_ATTR void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits)
{
bitD->bitsConsumed += nbBits;
}
/*! BIT_readBits() :
* Read (consume) next n bits from local register and update.
* Pay attention to not read more than nbBits contained into local register.
* @return : extracted value. */
MEM_STATIC FORCE_INLINE_ATTR size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits)
{
size_t const value = BIT_lookBits(bitD, nbBits);
BIT_skipBits(bitD, nbBits);
return value;
}
/*! BIT_readBitsFast() :
* unsafe version; only works if nbBits >= 1 */
MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits)
{
size_t const value = BIT_lookBitsFast(bitD, nbBits);
assert(nbBits >= 1);
BIT_skipBits(bitD, nbBits);
return value;
}
/*! BIT_reloadDStreamFast() :
* Similar to BIT_reloadDStream(), but with two differences:
* 1. bitsConsumed <= sizeof(bitD->bitContainer)*8 must hold!
* 2. Returns BIT_DStream_overflow when bitD->ptr < bitD->limitPtr, at this
* point you must use BIT_reloadDStream() to reload.
*/
MEM_STATIC BIT_DStream_status BIT_reloadDStreamFast(BIT_DStream_t* bitD)
{
if (UNLIKELY(bitD->ptr < bitD->limitPtr))
return BIT_DStream_overflow;
assert(bitD->bitsConsumed <= sizeof(bitD->bitContainer)*8);
bitD->ptr -= bitD->bitsConsumed >> 3;
bitD->bitsConsumed &= 7;
bitD->bitContainer = MEM_readLEST(bitD->ptr);
return BIT_DStream_unfinished;
}
/*! BIT_reloadDStream() :
* Refill `bitD` from buffer previously set in BIT_initDStream() .
* This function is safe, it guarantees it will not read beyond src buffer.
* @return : status of `BIT_DStream_t` internal register.
* when status == BIT_DStream_unfinished, internal register is filled with at least 25 or 57 bits */
MEM_STATIC FORCE_INLINE_ATTR BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD)
{
if (bitD->bitsConsumed > (sizeof(bitD->bitContainer)*8)) /* overflow detected, like end of stream */
return BIT_DStream_overflow;
if (bitD->ptr >= bitD->limitPtr) {
return BIT_reloadDStreamFast(bitD);
}
if (bitD->ptr == bitD->start) {
if (bitD->bitsConsumed < sizeof(bitD->bitContainer)*8) return BIT_DStream_endOfBuffer;
return BIT_DStream_completed;
}
/* start < ptr < limitPtr */
{ U32 nbBytes = bitD->bitsConsumed >> 3;
BIT_DStream_status result = BIT_DStream_unfinished;
if (bitD->ptr - nbBytes < bitD->start) {
nbBytes = (U32)(bitD->ptr - bitD->start); /* ptr > start */
result = BIT_DStream_endOfBuffer;
}
bitD->ptr -= nbBytes;
bitD->bitsConsumed -= nbBytes*8;
bitD->bitContainer = MEM_readLEST(bitD->ptr); /* reminder : srcSize > sizeof(bitD->bitContainer), otherwise bitD->ptr == bitD->start */
return result;
}
}
/*! BIT_endOfDStream() :
* @return : 1 if DStream has _exactly_ reached its end (all bits consumed).
*/
MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* DStream)
{
return ((DStream->ptr == DStream->start) && (DStream->bitsConsumed == sizeof(DStream->bitContainer)*8));
}
#if defined (__cplusplus)
}
#endif
#endif /* BITSTREAM_H_MODULE */

358
raytracer/nvpro_core/third_party/zstd/lib/common/compiler.h vendored Normal file
View file

@ -0,0 +1,358 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMPILER_H
#define ZSTD_COMPILER_H
#include "portability_macros.h"
/*-*******************************************************
* Compiler specifics
*********************************************************/
/* force inlining */
#if !defined(ZSTD_NO_INLINE)
#if (defined(__GNUC__) && !defined(__STRICT_ANSI__)) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L /* C99 */
# define INLINE_KEYWORD inline
#else
# define INLINE_KEYWORD
#endif
#if defined(__GNUC__) || defined(__ICCARM__)
# define FORCE_INLINE_ATTR __attribute__((always_inline))
#elif defined(_MSC_VER)
# define FORCE_INLINE_ATTR __forceinline
#else
# define FORCE_INLINE_ATTR
#endif
#else
#define INLINE_KEYWORD
#define FORCE_INLINE_ATTR
#endif
/**
On MSVC qsort requires that functions passed into it use the __cdecl calling conversion(CC).
This explicitly marks such functions as __cdecl so that the code will still compile
if a CC other than __cdecl has been made the default.
*/
#if defined(_MSC_VER)
# define WIN_CDECL __cdecl
#else
# define WIN_CDECL
#endif
/**
* FORCE_INLINE_TEMPLATE is used to define C "templates", which take constant
* parameters. They must be inlined for the compiler to eliminate the constant
* branches.
*/
#define FORCE_INLINE_TEMPLATE static INLINE_KEYWORD FORCE_INLINE_ATTR
/**
* HINT_INLINE is used to help the compiler generate better code. It is *not*
* used for "templates", so it can be tweaked based on the compilers
* performance.
*
* gcc-4.8 and gcc-4.9 have been shown to benefit from leaving off the
* always_inline attribute.
*
* clang up to 5.0.0 (trunk) benefit tremendously from the always_inline
* attribute.
*/
#if !defined(__clang__) && defined(__GNUC__) && __GNUC__ >= 4 && __GNUC_MINOR__ >= 8 && __GNUC__ < 5
# define HINT_INLINE static INLINE_KEYWORD
#else
# define HINT_INLINE static INLINE_KEYWORD FORCE_INLINE_ATTR
#endif
/* UNUSED_ATTR tells the compiler it is okay if the function is unused. */
#if defined(__GNUC__)
# define UNUSED_ATTR __attribute__((unused))
#else
# define UNUSED_ATTR
#endif
/* force no inlining */
#ifdef _MSC_VER
# define FORCE_NOINLINE static __declspec(noinline)
#else
# if defined(__GNUC__) || defined(__ICCARM__)
# define FORCE_NOINLINE static __attribute__((__noinline__))
# else
# define FORCE_NOINLINE static
# endif
#endif
/* target attribute */
#if defined(__GNUC__) || defined(__ICCARM__)
# define TARGET_ATTRIBUTE(target) __attribute__((__target__(target)))
#else
# define TARGET_ATTRIBUTE(target)
#endif
/* Target attribute for BMI2 dynamic dispatch.
* Enable lzcnt, bmi, and bmi2.
* We test for bmi1 & bmi2. lzcnt is included in bmi1.
*/
#define BMI2_TARGET_ATTRIBUTE TARGET_ATTRIBUTE("lzcnt,bmi,bmi2")
/* prefetch
* can be disabled, by declaring NO_PREFETCH build macro */
#if defined(NO_PREFETCH)
# define PREFETCH_L1(ptr) (void)(ptr) /* disabled */
# define PREFETCH_L2(ptr) (void)(ptr) /* disabled */
#else
# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86)) /* _mm_prefetch() is not defined outside of x86/x64 */
# include <mmintrin.h> /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */
# define PREFETCH_L1(ptr) _mm_prefetch((const char*)(ptr), _MM_HINT_T0)
# define PREFETCH_L2(ptr) _mm_prefetch((const char*)(ptr), _MM_HINT_T1)
# elif defined(__GNUC__) && ( (__GNUC__ >= 4) || ( (__GNUC__ == 3) && (__GNUC_MINOR__ >= 1) ) )
# define PREFETCH_L1(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 3 /* locality */)
# define PREFETCH_L2(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 2 /* locality */)
# elif defined(__aarch64__)
# define PREFETCH_L1(ptr) __asm__ __volatile__("prfm pldl1keep, %0" ::"Q"(*(ptr)))
# define PREFETCH_L2(ptr) __asm__ __volatile__("prfm pldl2keep, %0" ::"Q"(*(ptr)))
# else
# define PREFETCH_L1(ptr) (void)(ptr) /* disabled */
# define PREFETCH_L2(ptr) (void)(ptr) /* disabled */
# endif
#endif /* NO_PREFETCH */
#define CACHELINE_SIZE 64
#define PREFETCH_AREA(p, s) { \
const char* const _ptr = (const char*)(p); \
size_t const _size = (size_t)(s); \
size_t _pos; \
for (_pos=0; _pos<_size; _pos+=CACHELINE_SIZE) { \
PREFETCH_L2(_ptr + _pos); \
} \
}
/* vectorization
* older GCC (pre gcc-4.3 picked as the cutoff) uses a different syntax,
* and some compilers, like Intel ICC and MCST LCC, do not support it at all. */
#if !defined(__INTEL_COMPILER) && !defined(__clang__) && defined(__GNUC__) && !defined(__LCC__)
# if (__GNUC__ == 4 && __GNUC_MINOR__ > 3) || (__GNUC__ >= 5)
# define DONT_VECTORIZE __attribute__((optimize("no-tree-vectorize")))
# else
# define DONT_VECTORIZE _Pragma("GCC optimize(\"no-tree-vectorize\")")
# endif
#else
# define DONT_VECTORIZE
#endif
/* Tell the compiler that a branch is likely or unlikely.
* Only use these macros if it causes the compiler to generate better code.
* If you can remove a LIKELY/UNLIKELY annotation without speed changes in gcc
* and clang, please do.
*/
#if defined(__GNUC__)
#define LIKELY(x) (__builtin_expect((x), 1))
#define UNLIKELY(x) (__builtin_expect((x), 0))
#else
#define LIKELY(x) (x)
#define UNLIKELY(x) (x)
#endif
#if __has_builtin(__builtin_unreachable) || (defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 5)))
# define ZSTD_UNREACHABLE { assert(0), __builtin_unreachable(); }
#else
# define ZSTD_UNREACHABLE { assert(0); }
#endif
/* disable warnings */
#ifdef _MSC_VER /* Visual Studio */
# include <intrin.h> /* For Visual 2005 */
# pragma warning(disable : 4100) /* disable: C4100: unreferenced formal parameter */
# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
# pragma warning(disable : 4204) /* disable: C4204: non-constant aggregate initializer */
# pragma warning(disable : 4214) /* disable: C4214: non-int bitfields */
# pragma warning(disable : 4324) /* disable: C4324: padded structure */
#endif
/*Like DYNAMIC_BMI2 but for compile time determination of BMI2 support*/
#ifndef STATIC_BMI2
# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86))
# ifdef __AVX2__ //MSVC does not have a BMI2 specific flag, but every CPU that supports AVX2 also supports BMI2
# define STATIC_BMI2 1
# endif
# elif defined(__BMI2__) && defined(__x86_64__) && defined(__GNUC__)
# define STATIC_BMI2 1
# endif
#endif
#ifndef STATIC_BMI2
#define STATIC_BMI2 0
#endif
/* compile time determination of SIMD support */
#if !defined(ZSTD_NO_INTRINSICS)
# if defined(__SSE2__) || defined(_M_AMD64) || (defined (_M_IX86) && defined(_M_IX86_FP) && (_M_IX86_FP >= 2))
# define ZSTD_ARCH_X86_SSE2
# endif
# if defined(__ARM_NEON) || defined(_M_ARM64)
# define ZSTD_ARCH_ARM_NEON
# endif
#
# if defined(ZSTD_ARCH_X86_SSE2)
# include <emmintrin.h>
# elif defined(ZSTD_ARCH_ARM_NEON)
# include <arm_neon.h>
# endif
#endif
/* C-language Attributes are added in C23. */
#if defined(__STDC_VERSION__) && (__STDC_VERSION__ > 201710L) && defined(__has_c_attribute)
# define ZSTD_HAS_C_ATTRIBUTE(x) __has_c_attribute(x)
#else
# define ZSTD_HAS_C_ATTRIBUTE(x) 0
#endif
/* Only use C++ attributes in C++. Some compilers report support for C++
* attributes when compiling with C.
*/
#if defined(__cplusplus) && defined(__has_cpp_attribute)
# define ZSTD_HAS_CPP_ATTRIBUTE(x) __has_cpp_attribute(x)
#else
# define ZSTD_HAS_CPP_ATTRIBUTE(x) 0
#endif
/* Define ZSTD_FALLTHROUGH macro for annotating switch case with the 'fallthrough' attribute.
* - C23: https://en.cppreference.com/w/c/language/attributes/fallthrough
* - CPP17: https://en.cppreference.com/w/cpp/language/attributes/fallthrough
* - Else: __attribute__((__fallthrough__))
*/
#ifndef ZSTD_FALLTHROUGH
# if ZSTD_HAS_C_ATTRIBUTE(fallthrough)
# define ZSTD_FALLTHROUGH [[fallthrough]]
# elif ZSTD_HAS_CPP_ATTRIBUTE(fallthrough)
# define ZSTD_FALLTHROUGH [[fallthrough]]
# elif __has_attribute(__fallthrough__)
/* Leading semicolon is to satisfy gcc-11 with -pedantic. Without the semicolon
* gcc complains about: a label can only be part of a statement and a declaration is not a statement.
*/
# define ZSTD_FALLTHROUGH ; __attribute__((__fallthrough__))
# else
# define ZSTD_FALLTHROUGH
# endif
#endif
/*-**************************************************************
* Alignment check
*****************************************************************/
/* this test was initially positioned in mem.h,
* but this file is removed (or replaced) for linux kernel
* so it's now hosted in compiler.h,
* which remains valid for both user & kernel spaces.
*/
#ifndef ZSTD_ALIGNOF
# if defined(__GNUC__) || defined(_MSC_VER)
/* covers gcc, clang & MSVC */
/* note : this section must come first, before C11,
* due to a limitation in the kernel source generator */
# define ZSTD_ALIGNOF(T) __alignof(T)
# elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L)
/* C11 support */
# include <stdalign.h>
# define ZSTD_ALIGNOF(T) alignof(T)
# else
/* No known support for alignof() - imperfect backup */
# define ZSTD_ALIGNOF(T) (sizeof(void*) < sizeof(T) ? sizeof(void*) : sizeof(T))
# endif
#endif /* ZSTD_ALIGNOF */
/*-**************************************************************
* Sanitizer
*****************************************************************/
/* Issue #3240 reports an ASAN failure on an llvm-mingw build. Out of an
* abundance of caution, disable our custom poisoning on mingw. */
#ifdef __MINGW32__
#ifndef ZSTD_ASAN_DONT_POISON_WORKSPACE
#define ZSTD_ASAN_DONT_POISON_WORKSPACE 1
#endif
#ifndef ZSTD_MSAN_DONT_POISON_WORKSPACE
#define ZSTD_MSAN_DONT_POISON_WORKSPACE 1
#endif
#endif
#if ZSTD_MEMORY_SANITIZER && !defined(ZSTD_MSAN_DONT_POISON_WORKSPACE)
/* Not all platforms that support msan provide sanitizers/msan_interface.h.
* We therefore declare the functions we need ourselves, rather than trying to
* include the header file... */
#include <stddef.h> /* size_t */
#define ZSTD_DEPS_NEED_STDINT
#include "zstd_deps.h" /* intptr_t */
/* Make memory region fully initialized (without changing its contents). */
void __msan_unpoison(const volatile void *a, size_t size);
/* Make memory region fully uninitialized (without changing its contents).
This is a legacy interface that does not update origin information. Use
__msan_allocated_memory() instead. */
void __msan_poison(const volatile void *a, size_t size);
/* Returns the offset of the first (at least partially) poisoned byte in the
memory range, or -1 if the whole range is good. */
intptr_t __msan_test_shadow(const volatile void *x, size_t size);
/* Print shadow and origin for the memory range to stderr in a human-readable
format. */
void __msan_print_shadow(const volatile void *x, size_t size);
#endif
#if ZSTD_ADDRESS_SANITIZER && !defined(ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* Not all platforms that support asan provide sanitizers/asan_interface.h.
* We therefore declare the functions we need ourselves, rather than trying to
* include the header file... */
#include <stddef.h> /* size_t */
/**
* Marks a memory region (<c>[addr, addr+size)</c>) as unaddressable.
*
* This memory must be previously allocated by your program. Instrumented
* code is forbidden from accessing addresses in this region until it is
* unpoisoned. This function is not guaranteed to poison the entire region -
* it could poison only a subregion of <c>[addr, addr+size)</c> due to ASan
* alignment restrictions.
*
* \note This function is not thread-safe because no two threads can poison or
* unpoison memory in the same memory region simultaneously.
*
* \param addr Start of memory region.
* \param size Size of memory region. */
void __asan_poison_memory_region(void const volatile *addr, size_t size);
/**
* Marks a memory region (<c>[addr, addr+size)</c>) as addressable.
*
* This memory must be previously allocated by your program. Accessing
* addresses in this region is allowed until this region is poisoned again.
* This function could unpoison a super-region of <c>[addr, addr+size)</c> due
* to ASan alignment restrictions.
*
* \note This function is not thread-safe because no two threads can
* poison or unpoison memory in the same memory region simultaneously.
*
* \param addr Start of memory region.
* \param size Size of memory region. */
void __asan_unpoison_memory_region(void const volatile *addr, size_t size);
#endif
#endif /* ZSTD_COMPILER_H */

213
raytracer/nvpro_core/third_party/zstd/lib/common/cpu.h vendored Normal file
View file

@ -0,0 +1,213 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMMON_CPU_H
#define ZSTD_COMMON_CPU_H
/**
* Implementation taken from folly/CpuId.h
* https://github.com/facebook/folly/blob/master/folly/CpuId.h
*/
#include "mem.h"
#ifdef _MSC_VER
#include <intrin.h>
#endif
typedef struct {
U32 f1c;
U32 f1d;
U32 f7b;
U32 f7c;
} ZSTD_cpuid_t;
MEM_STATIC ZSTD_cpuid_t ZSTD_cpuid(void) {
U32 f1c = 0;
U32 f1d = 0;
U32 f7b = 0;
U32 f7c = 0;
#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_IX86))
int reg[4];
__cpuid((int*)reg, 0);
{
int const n = reg[0];
if (n >= 1) {
__cpuid((int*)reg, 1);
f1c = (U32)reg[2];
f1d = (U32)reg[3];
}
if (n >= 7) {
__cpuidex((int*)reg, 7, 0);
f7b = (U32)reg[1];
f7c = (U32)reg[2];
}
}
#elif defined(__i386__) && defined(__PIC__) && !defined(__clang__) && defined(__GNUC__)
/* The following block like the normal cpuid branch below, but gcc
* reserves ebx for use of its pic register so we must specially
* handle the save and restore to avoid clobbering the register
*/
U32 n;
__asm__(
"pushl %%ebx\n\t"
"cpuid\n\t"
"popl %%ebx\n\t"
: "=a"(n)
: "a"(0)
: "ecx", "edx");
if (n >= 1) {
U32 f1a;
__asm__(
"pushl %%ebx\n\t"
"cpuid\n\t"
"popl %%ebx\n\t"
: "=a"(f1a), "=c"(f1c), "=d"(f1d)
: "a"(1));
}
if (n >= 7) {
__asm__(
"pushl %%ebx\n\t"
"cpuid\n\t"
"movl %%ebx, %%eax\n\t"
"popl %%ebx"
: "=a"(f7b), "=c"(f7c)
: "a"(7), "c"(0)
: "edx");
}
#elif defined(__x86_64__) || defined(_M_X64) || defined(__i386__)
U32 n;
__asm__("cpuid" : "=a"(n) : "a"(0) : "ebx", "ecx", "edx");
if (n >= 1) {
U32 f1a;
__asm__("cpuid" : "=a"(f1a), "=c"(f1c), "=d"(f1d) : "a"(1) : "ebx");
}
if (n >= 7) {
U32 f7a;
__asm__("cpuid"
: "=a"(f7a), "=b"(f7b), "=c"(f7c)
: "a"(7), "c"(0)
: "edx");
}
#endif
{
ZSTD_cpuid_t cpuid;
cpuid.f1c = f1c;
cpuid.f1d = f1d;
cpuid.f7b = f7b;
cpuid.f7c = f7c;
return cpuid;
}
}
#define X(name, r, bit) \
MEM_STATIC int ZSTD_cpuid_##name(ZSTD_cpuid_t const cpuid) { \
return ((cpuid.r) & (1U << bit)) != 0; \
}
/* cpuid(1): Processor Info and Feature Bits. */
#define C(name, bit) X(name, f1c, bit)
C(sse3, 0)
C(pclmuldq, 1)
C(dtes64, 2)
C(monitor, 3)
C(dscpl, 4)
C(vmx, 5)
C(smx, 6)
C(eist, 7)
C(tm2, 8)
C(ssse3, 9)
C(cnxtid, 10)
C(fma, 12)
C(cx16, 13)
C(xtpr, 14)
C(pdcm, 15)
C(pcid, 17)
C(dca, 18)
C(sse41, 19)
C(sse42, 20)
C(x2apic, 21)
C(movbe, 22)
C(popcnt, 23)
C(tscdeadline, 24)
C(aes, 25)
C(xsave, 26)
C(osxsave, 27)
C(avx, 28)
C(f16c, 29)
C(rdrand, 30)
#undef C
#define D(name, bit) X(name, f1d, bit)
D(fpu, 0)
D(vme, 1)
D(de, 2)
D(pse, 3)
D(tsc, 4)
D(msr, 5)
D(pae, 6)
D(mce, 7)
D(cx8, 8)
D(apic, 9)
D(sep, 11)
D(mtrr, 12)
D(pge, 13)
D(mca, 14)
D(cmov, 15)
D(pat, 16)
D(pse36, 17)
D(psn, 18)
D(clfsh, 19)
D(ds, 21)
D(acpi, 22)
D(mmx, 23)
D(fxsr, 24)
D(sse, 25)
D(sse2, 26)
D(ss, 27)
D(htt, 28)
D(tm, 29)
D(pbe, 31)
#undef D
/* cpuid(7): Extended Features. */
#define B(name, bit) X(name, f7b, bit)
B(bmi1, 3)
B(hle, 4)
B(avx2, 5)
B(smep, 7)
B(bmi2, 8)
B(erms, 9)
B(invpcid, 10)
B(rtm, 11)
B(mpx, 14)
B(avx512f, 16)
B(avx512dq, 17)
B(rdseed, 18)
B(adx, 19)
B(smap, 20)
B(avx512ifma, 21)
B(pcommit, 22)
B(clflushopt, 23)
B(clwb, 24)
B(avx512pf, 26)
B(avx512er, 27)
B(avx512cd, 28)
B(sha, 29)
B(avx512bw, 30)
B(avx512vl, 31)
#undef B
#define C(name, bit) X(name, f7c, bit)
C(prefetchwt1, 0)
C(avx512vbmi, 1)
#undef C
#undef X
#endif /* ZSTD_COMMON_CPU_H */

24
raytracer/nvpro_core/third_party/zstd/lib/common/debug.c vendored Normal file
View file

@ -0,0 +1,24 @@
/* ******************************************************************
* debug
* Part of FSE library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/*
* This module only hosts one global variable
* which can be used to dynamically influence the verbosity of traces,
* such as DEBUGLOG and RAWLOG
*/
#include "debug.h"
int g_debuglevel = DEBUGLEVEL;

107
raytracer/nvpro_core/third_party/zstd/lib/common/debug.h vendored Normal file
View file

@ -0,0 +1,107 @@
/* ******************************************************************
* debug
* Part of FSE library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/*
* The purpose of this header is to enable debug functions.
* They regroup assert(), DEBUGLOG() and RAWLOG() for run-time,
* and DEBUG_STATIC_ASSERT() for compile-time.
*
* By default, DEBUGLEVEL==0, which means run-time debug is disabled.
*
* Level 1 enables assert() only.
* Starting level 2, traces can be generated and pushed to stderr.
* The higher the level, the more verbose the traces.
*
* It's possible to dynamically adjust level using variable g_debug_level,
* which is only declared if DEBUGLEVEL>=2,
* and is a global variable, not multi-thread protected (use with care)
*/
#ifndef DEBUG_H_12987983217
#define DEBUG_H_12987983217
#if defined (__cplusplus)
extern "C" {
#endif
/* static assert is triggered at compile time, leaving no runtime artefact.
* static assert only works with compile-time constants.
* Also, this variant can only be used inside a function. */
#define DEBUG_STATIC_ASSERT(c) (void)sizeof(char[(c) ? 1 : -1])
/* DEBUGLEVEL is expected to be defined externally,
* typically through compiler command line.
* Value must be a number. */
#ifndef DEBUGLEVEL
# define DEBUGLEVEL 0
#endif
/* recommended values for DEBUGLEVEL :
* 0 : release mode, no debug, all run-time checks disabled
* 1 : enables assert() only, no display
* 2 : reserved, for currently active debug path
* 3 : events once per object lifetime (CCtx, CDict, etc.)
* 4 : events once per frame
* 5 : events once per block
* 6 : events once per sequence (verbose)
* 7+: events at every position (*very* verbose)
*
* It's generally inconvenient to output traces > 5.
* In which case, it's possible to selectively trigger high verbosity levels
* by modifying g_debug_level.
*/
#if (DEBUGLEVEL>=1)
# define ZSTD_DEPS_NEED_ASSERT
# include "zstd_deps.h"
#else
# ifndef assert /* assert may be already defined, due to prior #include <assert.h> */
# define assert(condition) ((void)0) /* disable assert (default) */
# endif
#endif
#if (DEBUGLEVEL>=2)
# define ZSTD_DEPS_NEED_IO
# include "zstd_deps.h"
extern int g_debuglevel; /* the variable is only declared,
it actually lives in debug.c,
and is shared by the whole process.
It's not thread-safe.
It's useful when enabling very verbose levels
on selective conditions (such as position in src) */
# define RAWLOG(l, ...) { \
if (l<=g_debuglevel) { \
ZSTD_DEBUG_PRINT(__VA_ARGS__); \
} }
# define DEBUGLOG(l, ...) { \
if (l<=g_debuglevel) { \
ZSTD_DEBUG_PRINT(__FILE__ ": " __VA_ARGS__); \
ZSTD_DEBUG_PRINT(" \n"); \
} }
#else
# define RAWLOG(l, ...) {} /* disabled */
# define DEBUGLOG(l, ...) {} /* disabled */
#endif
#if defined (__cplusplus)
}
#endif
#endif /* DEBUG_H_12987983217 */

340
raytracer/nvpro_core/third_party/zstd/lib/common/entropy_common.c vendored Normal file
View file

@ -0,0 +1,340 @@
/* ******************************************************************
* Common functions of New Generation Entropy library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* *************************************
* Dependencies
***************************************/
#include "mem.h"
#include "error_private.h" /* ERR_*, ERROR */
#define FSE_STATIC_LINKING_ONLY /* FSE_MIN_TABLELOG */
#include "fse.h"
#include "huf.h"
#include "bits.h" /* ZSDT_highbit32, ZSTD_countTrailingZeros32 */
/*=== Version ===*/
unsigned FSE_versionNumber(void) { return FSE_VERSION_NUMBER; }
/*=== Error Management ===*/
unsigned FSE_isError(size_t code) { return ERR_isError(code); }
const char* FSE_getErrorName(size_t code) { return ERR_getErrorName(code); }
unsigned HUF_isError(size_t code) { return ERR_isError(code); }
const char* HUF_getErrorName(size_t code) { return ERR_getErrorName(code); }
/*-**************************************************************
* FSE NCount encoding-decoding
****************************************************************/
FORCE_INLINE_TEMPLATE
size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
{
const BYTE* const istart = (const BYTE*) headerBuffer;
const BYTE* const iend = istart + hbSize;
const BYTE* ip = istart;
int nbBits;
int remaining;
int threshold;
U32 bitStream;
int bitCount;
unsigned charnum = 0;
unsigned const maxSV1 = *maxSVPtr + 1;
int previous0 = 0;
if (hbSize < 8) {
/* This function only works when hbSize >= 8 */
char buffer[8] = {0};
ZSTD_memcpy(buffer, headerBuffer, hbSize);
{ size_t const countSize = FSE_readNCount(normalizedCounter, maxSVPtr, tableLogPtr,
buffer, sizeof(buffer));
if (FSE_isError(countSize)) return countSize;
if (countSize > hbSize) return ERROR(corruption_detected);
return countSize;
} }
assert(hbSize >= 8);
/* init */
ZSTD_memset(normalizedCounter, 0, (*maxSVPtr+1) * sizeof(normalizedCounter[0])); /* all symbols not present in NCount have a frequency of 0 */
bitStream = MEM_readLE32(ip);
nbBits = (bitStream & 0xF) + FSE_MIN_TABLELOG; /* extract tableLog */
if (nbBits > FSE_TABLELOG_ABSOLUTE_MAX) return ERROR(tableLog_tooLarge);
bitStream >>= 4;
bitCount = 4;
*tableLogPtr = nbBits;
remaining = (1<<nbBits)+1;
threshold = 1<<nbBits;
nbBits++;
for (;;) {
if (previous0) {
/* Count the number of repeats. Each time the
* 2-bit repeat code is 0b11 there is another
* repeat.
* Avoid UB by setting the high bit to 1.
*/
int repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
while (repeats >= 12) {
charnum += 3 * 12;
if (LIKELY(ip <= iend-7)) {
ip += 3;
} else {
bitCount -= (int)(8 * (iend - 7 - ip));
bitCount &= 31;
ip = iend - 4;
}
bitStream = MEM_readLE32(ip) >> bitCount;
repeats = ZSTD_countTrailingZeros32(~bitStream | 0x80000000) >> 1;
}
charnum += 3 * repeats;
bitStream >>= 2 * repeats;
bitCount += 2 * repeats;
/* Add the final repeat which isn't 0b11. */
assert((bitStream & 3) < 3);
charnum += bitStream & 3;
bitCount += 2;
/* This is an error, but break and return an error
* at the end, because returning out of a loop makes
* it harder for the compiler to optimize.
*/
if (charnum >= maxSV1) break;
/* We don't need to set the normalized count to 0
* because we already memset the whole buffer to 0.
*/
if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
assert((bitCount >> 3) <= 3); /* For first condition to work */
ip += bitCount>>3;
bitCount &= 7;
} else {
bitCount -= (int)(8 * (iend - 4 - ip));
bitCount &= 31;
ip = iend - 4;
}
bitStream = MEM_readLE32(ip) >> bitCount;
}
{
int const max = (2*threshold-1) - remaining;
int count;
if ((bitStream & (threshold-1)) < (U32)max) {
count = bitStream & (threshold-1);
bitCount += nbBits-1;
} else {
count = bitStream & (2*threshold-1);
if (count >= threshold) count -= max;
bitCount += nbBits;
}
count--; /* extra accuracy */
/* When it matters (small blocks), this is a
* predictable branch, because we don't use -1.
*/
if (count >= 0) {
remaining -= count;
} else {
assert(count == -1);
remaining += count;
}
normalizedCounter[charnum++] = (short)count;
previous0 = !count;
assert(threshold > 1);
if (remaining < threshold) {
/* This branch can be folded into the
* threshold update condition because we
* know that threshold > 1.
*/
if (remaining <= 1) break;
nbBits = ZSTD_highbit32(remaining) + 1;
threshold = 1 << (nbBits - 1);
}
if (charnum >= maxSV1) break;
if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) {
ip += bitCount>>3;
bitCount &= 7;
} else {
bitCount -= (int)(8 * (iend - 4 - ip));
bitCount &= 31;
ip = iend - 4;
}
bitStream = MEM_readLE32(ip) >> bitCount;
} }
if (remaining != 1) return ERROR(corruption_detected);
/* Only possible when there are too many zeros. */
if (charnum > maxSV1) return ERROR(maxSymbolValue_tooSmall);
if (bitCount > 32) return ERROR(corruption_detected);
*maxSVPtr = charnum-1;
ip += (bitCount+7)>>3;
return ip-istart;
}
/* Avoids the FORCE_INLINE of the _body() function. */
static size_t FSE_readNCount_body_default(
short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
{
return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
}
#if DYNAMIC_BMI2
BMI2_TARGET_ATTRIBUTE static size_t FSE_readNCount_body_bmi2(
short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
{
return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
}
#endif
size_t FSE_readNCount_bmi2(
short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize, int bmi2)
{
#if DYNAMIC_BMI2
if (bmi2) {
return FSE_readNCount_body_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
}
#endif
(void)bmi2;
return FSE_readNCount_body_default(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize);
}
size_t FSE_readNCount(
short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr,
const void* headerBuffer, size_t hbSize)
{
return FSE_readNCount_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize, /* bmi2 */ 0);
}
/*! HUF_readStats() :
Read compact Huffman tree, saved by HUF_writeCTable().
`huffWeight` is destination buffer.
`rankStats` is assumed to be a table of at least HUF_TABLELOG_MAX U32.
@return : size read from `src` , or an error Code .
Note : Needed by HUF_readCTable() and HUF_readDTableX?() .
*/
size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize)
{
U32 wksp[HUF_READ_STATS_WORKSPACE_SIZE_U32];
return HUF_readStats_wksp(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, wksp, sizeof(wksp), /* flags */ 0);
}
FORCE_INLINE_TEMPLATE size_t
HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workSpace, size_t wkspSize,
int bmi2)
{
U32 weightTotal;
const BYTE* ip = (const BYTE*) src;
size_t iSize;
size_t oSize;
if (!srcSize) return ERROR(srcSize_wrong);
iSize = ip[0];
/* ZSTD_memset(huffWeight, 0, hwSize); *//* is not necessary, even though some analyzer complain ... */
if (iSize >= 128) { /* special header */
oSize = iSize - 127;
iSize = ((oSize+1)/2);
if (iSize+1 > srcSize) return ERROR(srcSize_wrong);
if (oSize >= hwSize) return ERROR(corruption_detected);
ip += 1;
{ U32 n;
for (n=0; n<oSize; n+=2) {
huffWeight[n] = ip[n/2] >> 4;
huffWeight[n+1] = ip[n/2] & 15;
} } }
else { /* header compressed with FSE (normal case) */
if (iSize+1 > srcSize) return ERROR(srcSize_wrong);
/* max (hwSize-1) values decoded, as last one is implied */
oSize = FSE_decompress_wksp_bmi2(huffWeight, hwSize-1, ip+1, iSize, 6, workSpace, wkspSize, bmi2);
if (FSE_isError(oSize)) return oSize;
}
/* collect weight stats */
ZSTD_memset(rankStats, 0, (HUF_TABLELOG_MAX + 1) * sizeof(U32));
weightTotal = 0;
{ U32 n; for (n=0; n<oSize; n++) {
if (huffWeight[n] > HUF_TABLELOG_MAX) return ERROR(corruption_detected);
rankStats[huffWeight[n]]++;
weightTotal += (1 << huffWeight[n]) >> 1;
} }
if (weightTotal == 0) return ERROR(corruption_detected);
/* get last non-null symbol weight (implied, total must be 2^n) */
{ U32 const tableLog = ZSTD_highbit32(weightTotal) + 1;
if (tableLog > HUF_TABLELOG_MAX) return ERROR(corruption_detected);
*tableLogPtr = tableLog;
/* determine last weight */
{ U32 const total = 1 << tableLog;
U32 const rest = total - weightTotal;
U32 const verif = 1 << ZSTD_highbit32(rest);
U32 const lastWeight = ZSTD_highbit32(rest) + 1;
if (verif != rest) return ERROR(corruption_detected); /* last value must be a clean power of 2 */
huffWeight[oSize] = (BYTE)lastWeight;
rankStats[lastWeight]++;
} }
/* check tree construction validity */
if ((rankStats[1] < 2) || (rankStats[1] & 1)) return ERROR(corruption_detected); /* by construction : at least 2 elts of rank 1, must be even */
/* results */
*nbSymbolsPtr = (U32)(oSize+1);
return iSize+1;
}
/* Avoids the FORCE_INLINE of the _body() function. */
static size_t HUF_readStats_body_default(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workSpace, size_t wkspSize)
{
return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 0);
}
#if DYNAMIC_BMI2
static BMI2_TARGET_ATTRIBUTE size_t HUF_readStats_body_bmi2(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workSpace, size_t wkspSize)
{
return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 1);
}
#endif
size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, U32* rankStats,
U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workSpace, size_t wkspSize,
int flags)
{
#if DYNAMIC_BMI2
if (flags & HUF_flags_bmi2) {
return HUF_readStats_body_bmi2(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
}
#endif
(void)flags;
return HUF_readStats_body_default(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize);
}

63
raytracer/nvpro_core/third_party/zstd/lib/common/error_private.c vendored Normal file
View file

@ -0,0 +1,63 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* The purpose of this file is to have a single list of error strings embedded in binary */
#include "error_private.h"
const char* ERR_getErrorString(ERR_enum code)
{
#ifdef ZSTD_STRIP_ERROR_STRINGS
(void)code;
return "Error strings stripped";
#else
static const char* const notErrorCode = "Unspecified error code";
switch( code )
{
case PREFIX(no_error): return "No error detected";
case PREFIX(GENERIC): return "Error (generic)";
case PREFIX(prefix_unknown): return "Unknown frame descriptor";
case PREFIX(version_unsupported): return "Version not supported";
case PREFIX(frameParameter_unsupported): return "Unsupported frame parameter";
case PREFIX(frameParameter_windowTooLarge): return "Frame requires too much memory for decoding";
case PREFIX(corruption_detected): return "Data corruption detected";
case PREFIX(checksum_wrong): return "Restored data doesn't match checksum";
case PREFIX(literals_headerWrong): return "Header of Literals' block doesn't respect format specification";
case PREFIX(parameter_unsupported): return "Unsupported parameter";
case PREFIX(parameter_combination_unsupported): return "Unsupported combination of parameters";
case PREFIX(parameter_outOfBound): return "Parameter is out of bound";
case PREFIX(init_missing): return "Context should be init first";
case PREFIX(memory_allocation): return "Allocation error : not enough memory";
case PREFIX(workSpace_tooSmall): return "workSpace buffer is not large enough";
case PREFIX(stage_wrong): return "Operation not authorized at current processing stage";
case PREFIX(tableLog_tooLarge): return "tableLog requires too much memory : unsupported";
case PREFIX(maxSymbolValue_tooLarge): return "Unsupported max Symbol Value : too large";
case PREFIX(maxSymbolValue_tooSmall): return "Specified maxSymbolValue is too small";
case PREFIX(stabilityCondition_notRespected): return "pledged buffer stability condition is not respected";
case PREFIX(dictionary_corrupted): return "Dictionary is corrupted";
case PREFIX(dictionary_wrong): return "Dictionary mismatch";
case PREFIX(dictionaryCreation_failed): return "Cannot create Dictionary from provided samples";
case PREFIX(dstSize_tooSmall): return "Destination buffer is too small";
case PREFIX(srcSize_wrong): return "Src size is incorrect";
case PREFIX(dstBuffer_null): return "Operation on NULL destination buffer";
case PREFIX(noForwardProgress_destFull): return "Operation made no progress over multiple calls, due to output buffer being full";
case PREFIX(noForwardProgress_inputEmpty): return "Operation made no progress over multiple calls, due to input being empty";
/* following error codes are not stable and may be removed or changed in a future version */
case PREFIX(frameIndex_tooLarge): return "Frame index is too large";
case PREFIX(seekableIO): return "An I/O error occurred when reading/seeking";
case PREFIX(dstBuffer_wrong): return "Destination buffer is wrong";
case PREFIX(srcBuffer_wrong): return "Source buffer is wrong";
case PREFIX(sequenceProducer_failed): return "Block-level external sequence producer returned an error code";
case PREFIX(externalSequences_invalid): return "External sequences are not valid";
case PREFIX(maxCode):
default: return notErrorCode;
}
#endif
}

159
raytracer/nvpro_core/third_party/zstd/lib/common/error_private.h vendored Normal file
View file

@ -0,0 +1,159 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* Note : this module is expected to remain private, do not expose it */
#ifndef ERROR_H_MODULE
#define ERROR_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/* ****************************************
* Dependencies
******************************************/
#include "../zstd_errors.h" /* enum list */
#include "compiler.h"
#include "debug.h"
#include "zstd_deps.h" /* size_t */
/* ****************************************
* Compiler-specific
******************************************/
#if defined(__GNUC__)
# define ERR_STATIC static __attribute__((unused))
#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
# define ERR_STATIC static inline
#elif defined(_MSC_VER)
# define ERR_STATIC static __inline
#else
# define ERR_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */
#endif
/*-****************************************
* Customization (error_public.h)
******************************************/
typedef ZSTD_ErrorCode ERR_enum;
#define PREFIX(name) ZSTD_error_##name
/*-****************************************
* Error codes handling
******************************************/
#undef ERROR /* already defined on Visual Studio */
#define ERROR(name) ZSTD_ERROR(name)
#define ZSTD_ERROR(name) ((size_t)-PREFIX(name))
ERR_STATIC unsigned ERR_isError(size_t code) { return (code > ERROR(maxCode)); }
ERR_STATIC ERR_enum ERR_getErrorCode(size_t code) { if (!ERR_isError(code)) return (ERR_enum)0; return (ERR_enum) (0-code); }
/* check and forward error code */
#define CHECK_V_F(e, f) size_t const e = f; if (ERR_isError(e)) return e
#define CHECK_F(f) { CHECK_V_F(_var_err__, f); }
/*-****************************************
* Error Strings
******************************************/
const char* ERR_getErrorString(ERR_enum code); /* error_private.c */
ERR_STATIC const char* ERR_getErrorName(size_t code)
{
return ERR_getErrorString(ERR_getErrorCode(code));
}
/**
* Ignore: this is an internal helper.
*
* This is a helper function to help force C99-correctness during compilation.
* Under strict compilation modes, variadic macro arguments can't be empty.
* However, variadic function arguments can be. Using a function therefore lets
* us statically check that at least one (string) argument was passed,
* independent of the compilation flags.
*/
static INLINE_KEYWORD UNUSED_ATTR
void _force_has_format_string(const char *format, ...) {
(void)format;
}
/**
* Ignore: this is an internal helper.
*
* We want to force this function invocation to be syntactically correct, but
* we don't want to force runtime evaluation of its arguments.
*/
#define _FORCE_HAS_FORMAT_STRING(...) \
if (0) { \
_force_has_format_string(__VA_ARGS__); \
}
#define ERR_QUOTE(str) #str
/**
* Return the specified error if the condition evaluates to true.
*
* In debug modes, prints additional information.
* In order to do that (particularly, printing the conditional that failed),
* this can't just wrap RETURN_ERROR().
*/
#define RETURN_ERROR_IF(cond, err, ...) \
if (cond) { \
RAWLOG(3, "%s:%d: ERROR!: check %s failed, returning %s", \
__FILE__, __LINE__, ERR_QUOTE(cond), ERR_QUOTE(ERROR(err))); \
_FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \
RAWLOG(3, ": " __VA_ARGS__); \
RAWLOG(3, "\n"); \
return ERROR(err); \
}
/**
* Unconditionally return the specified error.
*
* In debug modes, prints additional information.
*/
#define RETURN_ERROR(err, ...) \
do { \
RAWLOG(3, "%s:%d: ERROR!: unconditional check failed, returning %s", \
__FILE__, __LINE__, ERR_QUOTE(ERROR(err))); \
_FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \
RAWLOG(3, ": " __VA_ARGS__); \
RAWLOG(3, "\n"); \
return ERROR(err); \
} while(0);
/**
* If the provided expression evaluates to an error code, returns that error code.
*
* In debug modes, prints additional information.
*/
#define FORWARD_IF_ERROR(err, ...) \
do { \
size_t const err_code = (err); \
if (ERR_isError(err_code)) { \
RAWLOG(3, "%s:%d: ERROR!: forwarding error in %s: %s", \
__FILE__, __LINE__, ERR_QUOTE(err), ERR_getErrorName(err_code)); \
_FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \
RAWLOG(3, ": " __VA_ARGS__); \
RAWLOG(3, "\n"); \
return err_code; \
} \
} while(0);
#if defined (__cplusplus)
}
#endif
#endif /* ERROR_H_MODULE */

639
raytracer/nvpro_core/third_party/zstd/lib/common/fse.h vendored Normal file
View file

@ -0,0 +1,639 @@
/* ******************************************************************
* FSE : Finite State Entropy codec
* Public Prototypes declaration
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
#if defined (__cplusplus)
extern "C" {
#endif
#ifndef FSE_H
#define FSE_H
/*-*****************************************
* Dependencies
******************************************/
#include "zstd_deps.h" /* size_t, ptrdiff_t */
/*-*****************************************
* FSE_PUBLIC_API : control library symbols visibility
******************************************/
#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)
# define FSE_PUBLIC_API __attribute__ ((visibility ("default")))
#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) /* Visual expected */
# define FSE_PUBLIC_API __declspec(dllexport)
#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)
# define FSE_PUBLIC_API __declspec(dllimport) /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/
#else
# define FSE_PUBLIC_API
#endif
/*------ Version ------*/
#define FSE_VERSION_MAJOR 0
#define FSE_VERSION_MINOR 9
#define FSE_VERSION_RELEASE 0
#define FSE_LIB_VERSION FSE_VERSION_MAJOR.FSE_VERSION_MINOR.FSE_VERSION_RELEASE
#define FSE_QUOTE(str) #str
#define FSE_EXPAND_AND_QUOTE(str) FSE_QUOTE(str)
#define FSE_VERSION_STRING FSE_EXPAND_AND_QUOTE(FSE_LIB_VERSION)
#define FSE_VERSION_NUMBER (FSE_VERSION_MAJOR *100*100 + FSE_VERSION_MINOR *100 + FSE_VERSION_RELEASE)
FSE_PUBLIC_API unsigned FSE_versionNumber(void); /**< library version number; to be used when checking dll version */
/*-*****************************************
* Tool functions
******************************************/
FSE_PUBLIC_API size_t FSE_compressBound(size_t size); /* maximum compressed size */
/* Error Management */
FSE_PUBLIC_API unsigned FSE_isError(size_t code); /* tells if a return value is an error code */
FSE_PUBLIC_API const char* FSE_getErrorName(size_t code); /* provides error code string (useful for debugging) */
/*-*****************************************
* FSE detailed API
******************************************/
/*!
FSE_compress() does the following:
1. count symbol occurrence from source[] into table count[] (see hist.h)
2. normalize counters so that sum(count[]) == Power_of_2 (2^tableLog)
3. save normalized counters to memory buffer using writeNCount()
4. build encoding table 'CTable' from normalized counters
5. encode the data stream using encoding table 'CTable'
FSE_decompress() does the following:
1. read normalized counters with readNCount()
2. build decoding table 'DTable' from normalized counters
3. decode the data stream using decoding table 'DTable'
The following API allows targeting specific sub-functions for advanced tasks.
For example, it's possible to compress several blocks using the same 'CTable',
or to save and provide normalized distribution using external method.
*/
/* *** COMPRESSION *** */
/*! FSE_optimalTableLog():
dynamically downsize 'tableLog' when conditions are met.
It saves CPU time, by using smaller tables, while preserving or even improving compression ratio.
@return : recommended tableLog (necessarily <= 'maxTableLog') */
FSE_PUBLIC_API unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);
/*! FSE_normalizeCount():
normalize counts so that sum(count[]) == Power_of_2 (2^tableLog)
'normalizedCounter' is a table of short, of minimum size (maxSymbolValue+1).
useLowProbCount is a boolean parameter which trades off compressed size for
faster header decoding. When it is set to 1, the compressed data will be slightly
smaller. And when it is set to 0, FSE_readNCount() and FSE_buildDTable() will be
faster. If you are compressing a small amount of data (< 2 KB) then useLowProbCount=0
is a good default, since header deserialization makes a big speed difference.
Otherwise, useLowProbCount=1 is a good default, since the speed difference is small.
@return : tableLog,
or an errorCode, which can be tested using FSE_isError() */
FSE_PUBLIC_API size_t FSE_normalizeCount(short* normalizedCounter, unsigned tableLog,
const unsigned* count, size_t srcSize, unsigned maxSymbolValue, unsigned useLowProbCount);
/*! FSE_NCountWriteBound():
Provides the maximum possible size of an FSE normalized table, given 'maxSymbolValue' and 'tableLog'.
Typically useful for allocation purpose. */
FSE_PUBLIC_API size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog);
/*! FSE_writeNCount():
Compactly save 'normalizedCounter' into 'buffer'.
@return : size of the compressed table,
or an errorCode, which can be tested using FSE_isError(). */
FSE_PUBLIC_API size_t FSE_writeNCount (void* buffer, size_t bufferSize,
const short* normalizedCounter,
unsigned maxSymbolValue, unsigned tableLog);
/*! Constructor and Destructor of FSE_CTable.
Note that FSE_CTable size depends on 'tableLog' and 'maxSymbolValue' */
typedef unsigned FSE_CTable; /* don't allocate that. It's only meant to be more restrictive than void* */
/*! FSE_buildCTable():
Builds `ct`, which must be already allocated, using FSE_createCTable().
@return : 0, or an errorCode, which can be tested using FSE_isError() */
FSE_PUBLIC_API size_t FSE_buildCTable(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog);
/*! FSE_compress_usingCTable():
Compress `src` using `ct` into `dst` which must be already allocated.
@return : size of compressed data (<= `dstCapacity`),
or 0 if compressed data could not fit into `dst`,
or an errorCode, which can be tested using FSE_isError() */
FSE_PUBLIC_API size_t FSE_compress_usingCTable (void* dst, size_t dstCapacity, const void* src, size_t srcSize, const FSE_CTable* ct);
/*!
Tutorial :
----------
The first step is to count all symbols. FSE_count() does this job very fast.
Result will be saved into 'count', a table of unsigned int, which must be already allocated, and have 'maxSymbolValuePtr[0]+1' cells.
'src' is a table of bytes of size 'srcSize'. All values within 'src' MUST be <= maxSymbolValuePtr[0]
maxSymbolValuePtr[0] will be updated, with its real value (necessarily <= original value)
FSE_count() will return the number of occurrence of the most frequent symbol.
This can be used to know if there is a single symbol within 'src', and to quickly evaluate its compressibility.
If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()).
The next step is to normalize the frequencies.
FSE_normalizeCount() will ensure that sum of frequencies is == 2 ^'tableLog'.
It also guarantees a minimum of 1 to any Symbol with frequency >= 1.
You can use 'tableLog'==0 to mean "use default tableLog value".
If you are unsure of which tableLog value to use, you can ask FSE_optimalTableLog(),
which will provide the optimal valid tableLog given sourceSize, maxSymbolValue, and a user-defined maximum (0 means "default").
The result of FSE_normalizeCount() will be saved into a table,
called 'normalizedCounter', which is a table of signed short.
'normalizedCounter' must be already allocated, and have at least 'maxSymbolValue+1' cells.
The return value is tableLog if everything proceeded as expected.
It is 0 if there is a single symbol within distribution.
If there is an error (ex: invalid tableLog value), the function will return an ErrorCode (which can be tested using FSE_isError()).
'normalizedCounter' can be saved in a compact manner to a memory area using FSE_writeNCount().
'buffer' must be already allocated.
For guaranteed success, buffer size must be at least FSE_headerBound().
The result of the function is the number of bytes written into 'buffer'.
If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError(); ex : buffer size too small).
'normalizedCounter' can then be used to create the compression table 'CTable'.
The space required by 'CTable' must be already allocated, using FSE_createCTable().
You can then use FSE_buildCTable() to fill 'CTable'.
If there is an error, both functions will return an ErrorCode (which can be tested using FSE_isError()).
'CTable' can then be used to compress 'src', with FSE_compress_usingCTable().
Similar to FSE_count(), the convention is that 'src' is assumed to be a table of char of size 'srcSize'
The function returns the size of compressed data (without header), necessarily <= `dstCapacity`.
If it returns '0', compressed data could not fit into 'dst'.
If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()).
*/
/* *** DECOMPRESSION *** */
/*! FSE_readNCount():
Read compactly saved 'normalizedCounter' from 'rBuffer'.
@return : size read from 'rBuffer',
or an errorCode, which can be tested using FSE_isError().
maxSymbolValuePtr[0] and tableLogPtr[0] will also be updated with their respective values */
FSE_PUBLIC_API size_t FSE_readNCount (short* normalizedCounter,
unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
const void* rBuffer, size_t rBuffSize);
/*! FSE_readNCount_bmi2():
* Same as FSE_readNCount() but pass bmi2=1 when your CPU supports BMI2 and 0 otherwise.
*/
FSE_PUBLIC_API size_t FSE_readNCount_bmi2(short* normalizedCounter,
unsigned* maxSymbolValuePtr, unsigned* tableLogPtr,
const void* rBuffer, size_t rBuffSize, int bmi2);
typedef unsigned FSE_DTable; /* don't allocate that. It's just a way to be more restrictive than void* */
/*!
Tutorial :
----------
(Note : these functions only decompress FSE-compressed blocks.
If block is uncompressed, use memcpy() instead
If block is a single repeated byte, use memset() instead )
The first step is to obtain the normalized frequencies of symbols.
This can be performed by FSE_readNCount() if it was saved using FSE_writeNCount().
'normalizedCounter' must be already allocated, and have at least 'maxSymbolValuePtr[0]+1' cells of signed short.
In practice, that means it's necessary to know 'maxSymbolValue' beforehand,
or size the table to handle worst case situations (typically 256).
FSE_readNCount() will provide 'tableLog' and 'maxSymbolValue'.
The result of FSE_readNCount() is the number of bytes read from 'rBuffer'.
Note that 'rBufferSize' must be at least 4 bytes, even if useful information is less than that.
If there is an error, the function will return an error code, which can be tested using FSE_isError().
The next step is to build the decompression tables 'FSE_DTable' from 'normalizedCounter'.
This is performed by the function FSE_buildDTable().
The space required by 'FSE_DTable' must be already allocated using FSE_createDTable().
If there is an error, the function will return an error code, which can be tested using FSE_isError().
`FSE_DTable` can then be used to decompress `cSrc`, with FSE_decompress_usingDTable().
`cSrcSize` must be strictly correct, otherwise decompression will fail.
FSE_decompress_usingDTable() result will tell how many bytes were regenerated (<=`dstCapacity`).
If there is an error, the function will return an error code, which can be tested using FSE_isError(). (ex: dst buffer too small)
*/
#endif /* FSE_H */
#if defined(FSE_STATIC_LINKING_ONLY) && !defined(FSE_H_FSE_STATIC_LINKING_ONLY)
#define FSE_H_FSE_STATIC_LINKING_ONLY
/* *** Dependency *** */
#include "bitstream.h"
/* *****************************************
* Static allocation
*******************************************/
/* FSE buffer bounds */
#define FSE_NCOUNTBOUND 512
#define FSE_BLOCKBOUND(size) ((size) + ((size)>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */)
#define FSE_COMPRESSBOUND(size) (FSE_NCOUNTBOUND + FSE_BLOCKBOUND(size)) /* Macro version, useful for static allocation */
/* It is possible to statically allocate FSE CTable/DTable as a table of FSE_CTable/FSE_DTable using below macros */
#define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) (1 + (1<<((maxTableLog)-1)) + (((maxSymbolValue)+1)*2))
#define FSE_DTABLE_SIZE_U32(maxTableLog) (1 + (1<<(maxTableLog)))
/* or use the size to malloc() space directly. Pay attention to alignment restrictions though */
#define FSE_CTABLE_SIZE(maxTableLog, maxSymbolValue) (FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(FSE_CTable))
#define FSE_DTABLE_SIZE(maxTableLog) (FSE_DTABLE_SIZE_U32(maxTableLog) * sizeof(FSE_DTable))
/* *****************************************
* FSE advanced API
***************************************** */
unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus);
/**< same as FSE_optimalTableLog(), which used `minus==2` */
size_t FSE_buildCTable_rle (FSE_CTable* ct, unsigned char symbolValue);
/**< build a fake FSE_CTable, designed to compress always the same symbolValue */
/* FSE_buildCTable_wksp() :
* Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`).
* `wkspSize` must be >= `FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)` of `unsigned`.
* See FSE_buildCTable_wksp() for breakdown of workspace usage.
*/
#define FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog) (((maxSymbolValue + 2) + (1ull << (tableLog)))/2 + sizeof(U64)/sizeof(U32) /* additional 8 bytes for potential table overwrite */)
#define FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) (sizeof(unsigned) * FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog))
size_t FSE_buildCTable_wksp(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
#define FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) (sizeof(short) * (maxSymbolValue + 1) + (1ULL << maxTableLog) + 8)
#define FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ((FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) + sizeof(unsigned) - 1) / sizeof(unsigned))
FSE_PUBLIC_API size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize);
/**< Same as FSE_buildDTable(), using an externally allocated `workspace` produced with `FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxSymbolValue)` */
#define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + 1 + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1)
#define FSE_DECOMPRESS_WKSP_SIZE(maxTableLog, maxSymbolValue) (FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(unsigned))
size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2);
/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)`.
* Set bmi2 to 1 if your CPU supports BMI2 or 0 if it doesn't */
typedef enum {
FSE_repeat_none, /**< Cannot use the previous table */
FSE_repeat_check, /**< Can use the previous table but it must be checked */
FSE_repeat_valid /**< Can use the previous table and it is assumed to be valid */
} FSE_repeat;
/* *****************************************
* FSE symbol compression API
*******************************************/
/*!
This API consists of small unitary functions, which highly benefit from being inlined.
Hence their body are included in next section.
*/
typedef struct {
ptrdiff_t value;
const void* stateTable;
const void* symbolTT;
unsigned stateLog;
} FSE_CState_t;
static void FSE_initCState(FSE_CState_t* CStatePtr, const FSE_CTable* ct);
static void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* CStatePtr, unsigned symbol);
static void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* CStatePtr);
/**<
These functions are inner components of FSE_compress_usingCTable().
They allow the creation of custom streams, mixing multiple tables and bit sources.
A key property to keep in mind is that encoding and decoding are done **in reverse direction**.
So the first symbol you will encode is the last you will decode, like a LIFO stack.
You will need a few variables to track your CStream. They are :
FSE_CTable ct; // Provided by FSE_buildCTable()
BIT_CStream_t bitStream; // bitStream tracking structure
FSE_CState_t state; // State tracking structure (can have several)
The first thing to do is to init bitStream and state.
size_t errorCode = BIT_initCStream(&bitStream, dstBuffer, maxDstSize);
FSE_initCState(&state, ct);
Note that BIT_initCStream() can produce an error code, so its result should be tested, using FSE_isError();
You can then encode your input data, byte after byte.
FSE_encodeSymbol() outputs a maximum of 'tableLog' bits at a time.
Remember decoding will be done in reverse direction.
FSE_encodeByte(&bitStream, &state, symbol);
At any time, you can also add any bit sequence.
Note : maximum allowed nbBits is 25, for compatibility with 32-bits decoders
BIT_addBits(&bitStream, bitField, nbBits);
The above methods don't commit data to memory, they just store it into local register, for speed.
Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
Writing data to memory is a manual operation, performed by the flushBits function.
BIT_flushBits(&bitStream);
Your last FSE encoding operation shall be to flush your last state value(s).
FSE_flushState(&bitStream, &state);
Finally, you must close the bitStream.
The function returns the size of CStream in bytes.
If data couldn't fit into dstBuffer, it will return a 0 ( == not compressible)
If there is an error, it returns an errorCode (which can be tested using FSE_isError()).
size_t size = BIT_closeCStream(&bitStream);
*/
/* *****************************************
* FSE symbol decompression API
*******************************************/
typedef struct {
size_t state;
const void* table; /* precise table may vary, depending on U16 */
} FSE_DState_t;
static void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt);
static unsigned char FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD);
static unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr);
/**<
Let's now decompose FSE_decompress_usingDTable() into its unitary components.
You will decode FSE-encoded symbols from the bitStream,
and also any other bitFields you put in, **in reverse order**.
You will need a few variables to track your bitStream. They are :
BIT_DStream_t DStream; // Stream context
FSE_DState_t DState; // State context. Multiple ones are possible
FSE_DTable* DTablePtr; // Decoding table, provided by FSE_buildDTable()
The first thing to do is to init the bitStream.
errorCode = BIT_initDStream(&DStream, srcBuffer, srcSize);
You should then retrieve your initial state(s)
(in reverse flushing order if you have several ones) :
errorCode = FSE_initDState(&DState, &DStream, DTablePtr);
You can then decode your data, symbol after symbol.
For information the maximum number of bits read by FSE_decodeSymbol() is 'tableLog'.
Keep in mind that symbols are decoded in reverse order, like a LIFO stack (last in, first out).
unsigned char symbol = FSE_decodeSymbol(&DState, &DStream);
You can retrieve any bitfield you eventually stored into the bitStream (in reverse order)
Note : maximum allowed nbBits is 25, for 32-bits compatibility
size_t bitField = BIT_readBits(&DStream, nbBits);
All above operations only read from local register (which size depends on size_t).
Refueling the register from memory is manually performed by the reload method.
endSignal = FSE_reloadDStream(&DStream);
BIT_reloadDStream() result tells if there is still some more data to read from DStream.
BIT_DStream_unfinished : there is still some data left into the DStream.
BIT_DStream_endOfBuffer : Dstream reached end of buffer. Its container may no longer be completely filled.
BIT_DStream_completed : Dstream reached its exact end, corresponding in general to decompression completed.
BIT_DStream_tooFar : Dstream went too far. Decompression result is corrupted.
When reaching end of buffer (BIT_DStream_endOfBuffer), progress slowly, notably if you decode multiple symbols per loop,
to properly detect the exact end of stream.
After each decoded symbol, check if DStream is fully consumed using this simple test :
BIT_reloadDStream(&DStream) >= BIT_DStream_completed
When it's done, verify decompression is fully completed, by checking both DStream and the relevant states.
Checking if DStream has reached its end is performed by :
BIT_endOfDStream(&DStream);
Check also the states. There might be some symbols left there, if some high probability ones (>50%) are possible.
FSE_endOfDState(&DState);
*/
/* *****************************************
* FSE unsafe API
*******************************************/
static unsigned char FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD);
/* faster, but works only if nbBits is always >= 1 (otherwise, result will be corrupted) */
/* *****************************************
* Implementation of inlined functions
*******************************************/
typedef struct {
int deltaFindState;
U32 deltaNbBits;
} FSE_symbolCompressionTransform; /* total 8 bytes */
MEM_STATIC void FSE_initCState(FSE_CState_t* statePtr, const FSE_CTable* ct)
{
const void* ptr = ct;
const U16* u16ptr = (const U16*) ptr;
const U32 tableLog = MEM_read16(ptr);
statePtr->value = (ptrdiff_t)1<<tableLog;
statePtr->stateTable = u16ptr+2;
statePtr->symbolTT = ct + 1 + (tableLog ? (1<<(tableLog-1)) : 1);
statePtr->stateLog = tableLog;
}
/*! FSE_initCState2() :
* Same as FSE_initCState(), but the first symbol to include (which will be the last to be read)
* uses the smallest state value possible, saving the cost of this symbol */
MEM_STATIC void FSE_initCState2(FSE_CState_t* statePtr, const FSE_CTable* ct, U32 symbol)
{
FSE_initCState(statePtr, ct);
{ const FSE_symbolCompressionTransform symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol];
const U16* stateTable = (const U16*)(statePtr->stateTable);
U32 nbBitsOut = (U32)((symbolTT.deltaNbBits + (1<<15)) >> 16);
statePtr->value = (nbBitsOut << 16) - symbolTT.deltaNbBits;
statePtr->value = stateTable[(statePtr->value >> nbBitsOut) + symbolTT.deltaFindState];
}
}
MEM_STATIC void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* statePtr, unsigned symbol)
{
FSE_symbolCompressionTransform const symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol];
const U16* const stateTable = (const U16*)(statePtr->stateTable);
U32 const nbBitsOut = (U32)((statePtr->value + symbolTT.deltaNbBits) >> 16);
BIT_addBits(bitC, statePtr->value, nbBitsOut);
statePtr->value = stateTable[ (statePtr->value >> nbBitsOut) + symbolTT.deltaFindState];
}
MEM_STATIC void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* statePtr)
{
BIT_addBits(bitC, statePtr->value, statePtr->stateLog);
BIT_flushBits(bitC);
}
/* FSE_getMaxNbBits() :
* Approximate maximum cost of a symbol, in bits.
* Fractional get rounded up (i.e. a symbol with a normalized frequency of 3 gives the same result as a frequency of 2)
* note 1 : assume symbolValue is valid (<= maxSymbolValue)
* note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */
MEM_STATIC U32 FSE_getMaxNbBits(const void* symbolTTPtr, U32 symbolValue)
{
const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr;
return (symbolTT[symbolValue].deltaNbBits + ((1<<16)-1)) >> 16;
}
/* FSE_bitCost() :
* Approximate symbol cost, as fractional value, using fixed-point format (accuracyLog fractional bits)
* note 1 : assume symbolValue is valid (<= maxSymbolValue)
* note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */
MEM_STATIC U32 FSE_bitCost(const void* symbolTTPtr, U32 tableLog, U32 symbolValue, U32 accuracyLog)
{
const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr;
U32 const minNbBits = symbolTT[symbolValue].deltaNbBits >> 16;
U32 const threshold = (minNbBits+1) << 16;
assert(tableLog < 16);
assert(accuracyLog < 31-tableLog); /* ensure enough room for renormalization double shift */
{ U32 const tableSize = 1 << tableLog;
U32 const deltaFromThreshold = threshold - (symbolTT[symbolValue].deltaNbBits + tableSize);
U32 const normalizedDeltaFromThreshold = (deltaFromThreshold << accuracyLog) >> tableLog; /* linear interpolation (very approximate) */
U32 const bitMultiplier = 1 << accuracyLog;
assert(symbolTT[symbolValue].deltaNbBits + tableSize <= threshold);
assert(normalizedDeltaFromThreshold <= bitMultiplier);
return (minNbBits+1)*bitMultiplier - normalizedDeltaFromThreshold;
}
}
/* ====== Decompression ====== */
typedef struct {
U16 tableLog;
U16 fastMode;
} FSE_DTableHeader; /* sizeof U32 */
typedef struct
{
unsigned short newState;
unsigned char symbol;
unsigned char nbBits;
} FSE_decode_t; /* size == U32 */
MEM_STATIC void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt)
{
const void* ptr = dt;
const FSE_DTableHeader* const DTableH = (const FSE_DTableHeader*)ptr;
DStatePtr->state = BIT_readBits(bitD, DTableH->tableLog);
BIT_reloadDStream(bitD);
DStatePtr->table = dt + 1;
}
MEM_STATIC BYTE FSE_peekSymbol(const FSE_DState_t* DStatePtr)
{
FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
return DInfo.symbol;
}
MEM_STATIC void FSE_updateState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
{
FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
U32 const nbBits = DInfo.nbBits;
size_t const lowBits = BIT_readBits(bitD, nbBits);
DStatePtr->state = DInfo.newState + lowBits;
}
MEM_STATIC BYTE FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
{
FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
U32 const nbBits = DInfo.nbBits;
BYTE const symbol = DInfo.symbol;
size_t const lowBits = BIT_readBits(bitD, nbBits);
DStatePtr->state = DInfo.newState + lowBits;
return symbol;
}
/*! FSE_decodeSymbolFast() :
unsafe, only works if no symbol has a probability > 50% */
MEM_STATIC BYTE FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD)
{
FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state];
U32 const nbBits = DInfo.nbBits;
BYTE const symbol = DInfo.symbol;
size_t const lowBits = BIT_readBitsFast(bitD, nbBits);
DStatePtr->state = DInfo.newState + lowBits;
return symbol;
}
MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr)
{
return DStatePtr->state == 0;
}
#ifndef FSE_COMMONDEFS_ONLY
/* **************************************************************
* Tuning parameters
****************************************************************/
/*!MEMORY_USAGE :
* Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; etc.)
* Increasing memory usage improves compression ratio
* Reduced memory usage can improve speed, due to cache effect
* Recommended max value is 14, for 16KB, which nicely fits into Intel x86 L1 cache */
#ifndef FSE_MAX_MEMORY_USAGE
# define FSE_MAX_MEMORY_USAGE 14
#endif
#ifndef FSE_DEFAULT_MEMORY_USAGE
# define FSE_DEFAULT_MEMORY_USAGE 13
#endif
#if (FSE_DEFAULT_MEMORY_USAGE > FSE_MAX_MEMORY_USAGE)
# error "FSE_DEFAULT_MEMORY_USAGE must be <= FSE_MAX_MEMORY_USAGE"
#endif
/*!FSE_MAX_SYMBOL_VALUE :
* Maximum symbol value authorized.
* Required for proper stack allocation */
#ifndef FSE_MAX_SYMBOL_VALUE
# define FSE_MAX_SYMBOL_VALUE 255
#endif
/* **************************************************************
* template functions type & suffix
****************************************************************/
#define FSE_FUNCTION_TYPE BYTE
#define FSE_FUNCTION_EXTENSION
#define FSE_DECODE_TYPE FSE_decode_t
#endif /* !FSE_COMMONDEFS_ONLY */
/* ***************************************************************
* Constants
*****************************************************************/
#define FSE_MAX_TABLELOG (FSE_MAX_MEMORY_USAGE-2)
#define FSE_MAX_TABLESIZE (1U<<FSE_MAX_TABLELOG)
#define FSE_MAXTABLESIZE_MASK (FSE_MAX_TABLESIZE-1)
#define FSE_DEFAULT_TABLELOG (FSE_DEFAULT_MEMORY_USAGE-2)
#define FSE_MIN_TABLELOG 5
#define FSE_TABLELOG_ABSOLUTE_MAX 15
#if FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX
# error "FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX is not supported"
#endif
#define FSE_TABLESTEP(tableSize) (((tableSize)>>1) + ((tableSize)>>3) + 3)
#endif /* FSE_STATIC_LINKING_ONLY */
#if defined (__cplusplus)
}
#endif

311
raytracer/nvpro_core/third_party/zstd/lib/common/fse_decompress.c vendored Normal file
View file

@ -0,0 +1,311 @@
/* ******************************************************************
* FSE : Finite State Entropy decoder
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* **************************************************************
* Includes
****************************************************************/
#include "debug.h" /* assert */
#include "bitstream.h"
#include "compiler.h"
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
#include "error_private.h"
#define ZSTD_DEPS_NEED_MALLOC
#include "zstd_deps.h"
#include "bits.h" /* ZSTD_highbit32 */
/* **************************************************************
* Error Management
****************************************************************/
#define FSE_isError ERR_isError
#define FSE_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c) /* use only *after* variable declarations */
/* **************************************************************
* Templates
****************************************************************/
/*
designed to be included
for type-specific functions (template emulation in C)
Objective is to write these functions only once, for improved maintenance
*/
/* safety checks */
#ifndef FSE_FUNCTION_EXTENSION
# error "FSE_FUNCTION_EXTENSION must be defined"
#endif
#ifndef FSE_FUNCTION_TYPE
# error "FSE_FUNCTION_TYPE must be defined"
#endif
/* Function names */
#define FSE_CAT(X,Y) X##Y
#define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y)
#define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y)
static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
{
void* const tdPtr = dt+1; /* because *dt is unsigned, 32-bits aligned on 32-bits */
FSE_DECODE_TYPE* const tableDecode = (FSE_DECODE_TYPE*) (tdPtr);
U16* symbolNext = (U16*)workSpace;
BYTE* spread = (BYTE*)(symbolNext + maxSymbolValue + 1);
U32 const maxSV1 = maxSymbolValue + 1;
U32 const tableSize = 1 << tableLog;
U32 highThreshold = tableSize-1;
/* Sanity Checks */
if (FSE_BUILD_DTABLE_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(maxSymbolValue_tooLarge);
if (maxSymbolValue > FSE_MAX_SYMBOL_VALUE) return ERROR(maxSymbolValue_tooLarge);
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge);
/* Init, lay down lowprob symbols */
{ FSE_DTableHeader DTableH;
DTableH.tableLog = (U16)tableLog;
DTableH.fastMode = 1;
{ S16 const largeLimit= (S16)(1 << (tableLog-1));
U32 s;
for (s=0; s<maxSV1; s++) {
if (normalizedCounter[s]==-1) {
tableDecode[highThreshold--].symbol = (FSE_FUNCTION_TYPE)s;
symbolNext[s] = 1;
} else {
if (normalizedCounter[s] >= largeLimit) DTableH.fastMode=0;
symbolNext[s] = normalizedCounter[s];
} } }
ZSTD_memcpy(dt, &DTableH, sizeof(DTableH));
}
/* Spread symbols */
if (highThreshold == tableSize - 1) {
size_t const tableMask = tableSize-1;
size_t const step = FSE_TABLESTEP(tableSize);
/* First lay down the symbols in order.
* We use a uint64_t to lay down 8 bytes at a time. This reduces branch
* misses since small blocks generally have small table logs, so nearly
* all symbols have counts <= 8. We ensure we have 8 bytes at the end of
* our buffer to handle the over-write.
*/
{
U64 const add = 0x0101010101010101ull;
size_t pos = 0;
U64 sv = 0;
U32 s;
for (s=0; s<maxSV1; ++s, sv += add) {
int i;
int const n = normalizedCounter[s];
MEM_write64(spread + pos, sv);
for (i = 8; i < n; i += 8) {
MEM_write64(spread + pos + i, sv);
}
pos += n;
}
}
/* Now we spread those positions across the table.
* The benefit of doing it in two stages is that we avoid the
* variable size inner loop, which caused lots of branch misses.
* Now we can run through all the positions without any branch misses.
* We unroll the loop twice, since that is what empirically worked best.
*/
{
size_t position = 0;
size_t s;
size_t const unroll = 2;
assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */
for (s = 0; s < (size_t)tableSize; s += unroll) {
size_t u;
for (u = 0; u < unroll; ++u) {
size_t const uPosition = (position + (u * step)) & tableMask;
tableDecode[uPosition].symbol = spread[s + u];
}
position = (position + (unroll * step)) & tableMask;
}
assert(position == 0);
}
} else {
U32 const tableMask = tableSize-1;
U32 const step = FSE_TABLESTEP(tableSize);
U32 s, position = 0;
for (s=0; s<maxSV1; s++) {
int i;
for (i=0; i<normalizedCounter[s]; i++) {
tableDecode[position].symbol = (FSE_FUNCTION_TYPE)s;
position = (position + step) & tableMask;
while (position > highThreshold) position = (position + step) & tableMask; /* lowprob area */
} }
if (position!=0) return ERROR(GENERIC); /* position must reach all cells once, otherwise normalizedCounter is incorrect */
}
/* Build Decoding table */
{ U32 u;
for (u=0; u<tableSize; u++) {
FSE_FUNCTION_TYPE const symbol = (FSE_FUNCTION_TYPE)(tableDecode[u].symbol);
U32 const nextState = symbolNext[symbol]++;
tableDecode[u].nbBits = (BYTE) (tableLog - ZSTD_highbit32(nextState) );
tableDecode[u].newState = (U16) ( (nextState << tableDecode[u].nbBits) - tableSize);
} }
return 0;
}
size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize)
{
return FSE_buildDTable_internal(dt, normalizedCounter, maxSymbolValue, tableLog, workSpace, wkspSize);
}
#ifndef FSE_COMMONDEFS_ONLY
/*-*******************************************************
* Decompression (Byte symbols)
*********************************************************/
FORCE_INLINE_TEMPLATE size_t FSE_decompress_usingDTable_generic(
void* dst, size_t maxDstSize,
const void* cSrc, size_t cSrcSize,
const FSE_DTable* dt, const unsigned fast)
{
BYTE* const ostart = (BYTE*) dst;
BYTE* op = ostart;
BYTE* const omax = op + maxDstSize;
BYTE* const olimit = omax-3;
BIT_DStream_t bitD;
FSE_DState_t state1;
FSE_DState_t state2;
/* Init */
CHECK_F(BIT_initDStream(&bitD, cSrc, cSrcSize));
FSE_initDState(&state1, &bitD, dt);
FSE_initDState(&state2, &bitD, dt);
#define FSE_GETSYMBOL(statePtr) fast ? FSE_decodeSymbolFast(statePtr, &bitD) : FSE_decodeSymbol(statePtr, &bitD)
/* 4 symbols per loop */
for ( ; (BIT_reloadDStream(&bitD)==BIT_DStream_unfinished) & (op<olimit) ; op+=4) {
op[0] = FSE_GETSYMBOL(&state1);
if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */
BIT_reloadDStream(&bitD);
op[1] = FSE_GETSYMBOL(&state2);
if (FSE_MAX_TABLELOG*4+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */
{ if (BIT_reloadDStream(&bitD) > BIT_DStream_unfinished) { op+=2; break; } }
op[2] = FSE_GETSYMBOL(&state1);
if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */
BIT_reloadDStream(&bitD);
op[3] = FSE_GETSYMBOL(&state2);
}
/* tail */
/* note : BIT_reloadDStream(&bitD) >= FSE_DStream_partiallyFilled; Ends at exactly BIT_DStream_completed */
while (1) {
if (op>(omax-2)) return ERROR(dstSize_tooSmall);
*op++ = FSE_GETSYMBOL(&state1);
if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) {
*op++ = FSE_GETSYMBOL(&state2);
break;
}
if (op>(omax-2)) return ERROR(dstSize_tooSmall);
*op++ = FSE_GETSYMBOL(&state2);
if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) {
*op++ = FSE_GETSYMBOL(&state1);
break;
} }
return op-ostart;
}
typedef struct {
short ncount[FSE_MAX_SYMBOL_VALUE + 1];
FSE_DTable dtable[1]; /* Dynamically sized */
} FSE_DecompressWksp;
FORCE_INLINE_TEMPLATE size_t FSE_decompress_wksp_body(
void* dst, size_t dstCapacity,
const void* cSrc, size_t cSrcSize,
unsigned maxLog, void* workSpace, size_t wkspSize,
int bmi2)
{
const BYTE* const istart = (const BYTE*)cSrc;
const BYTE* ip = istart;
unsigned tableLog;
unsigned maxSymbolValue = FSE_MAX_SYMBOL_VALUE;
FSE_DecompressWksp* const wksp = (FSE_DecompressWksp*)workSpace;
DEBUG_STATIC_ASSERT((FSE_MAX_SYMBOL_VALUE + 1) % 2 == 0);
if (wkspSize < sizeof(*wksp)) return ERROR(GENERIC);
/* normal FSE decoding mode */
{
size_t const NCountLength = FSE_readNCount_bmi2(wksp->ncount, &maxSymbolValue, &tableLog, istart, cSrcSize, bmi2);
if (FSE_isError(NCountLength)) return NCountLength;
if (tableLog > maxLog) return ERROR(tableLog_tooLarge);
assert(NCountLength <= cSrcSize);
ip += NCountLength;
cSrcSize -= NCountLength;
}
if (FSE_DECOMPRESS_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(tableLog_tooLarge);
assert(sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog) <= wkspSize);
workSpace = (BYTE*)workSpace + sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
wkspSize -= sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog);
CHECK_F( FSE_buildDTable_internal(wksp->dtable, wksp->ncount, maxSymbolValue, tableLog, workSpace, wkspSize) );
{
const void* ptr = wksp->dtable;
const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr;
const U32 fastMode = DTableH->fastMode;
/* select fast mode (static) */
if (fastMode) return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 1);
return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 0);
}
}
/* Avoids the FORCE_INLINE of the _body() function. */
static size_t FSE_decompress_wksp_body_default(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
{
return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 0);
}
#if DYNAMIC_BMI2
BMI2_TARGET_ATTRIBUTE static size_t FSE_decompress_wksp_body_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize)
{
return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 1);
}
#endif
size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2)
{
#if DYNAMIC_BMI2
if (bmi2) {
return FSE_decompress_wksp_body_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize);
}
#endif
(void)bmi2;
return FSE_decompress_wksp_body_default(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize);
}
#endif /* FSE_COMMONDEFS_ONLY */

273
raytracer/nvpro_core/third_party/zstd/lib/common/huf.h vendored Normal file
View file

@ -0,0 +1,273 @@
/* ******************************************************************
* huff0 huffman codec,
* part of Finite State Entropy library
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
#if defined (__cplusplus)
extern "C" {
#endif
#ifndef HUF_H_298734234
#define HUF_H_298734234
/* *** Dependencies *** */
#include "zstd_deps.h" /* size_t */
#include "mem.h" /* U32 */
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
/* *** Tool functions *** */
#define HUF_BLOCKSIZE_MAX (128 * 1024) /**< maximum input size for a single block compressed with HUF_compress */
size_t HUF_compressBound(size_t size); /**< maximum compressed size (worst case) */
/* Error Management */
unsigned HUF_isError(size_t code); /**< tells if a return value is an error code */
const char* HUF_getErrorName(size_t code); /**< provides error code string (useful for debugging) */
#define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */)
#define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64))
/* *** Constants *** */
#define HUF_TABLELOG_MAX 12 /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */
#define HUF_TABLELOG_DEFAULT 11 /* default tableLog value when none specified */
#define HUF_SYMBOLVALUE_MAX 255
#define HUF_TABLELOG_ABSOLUTEMAX 12 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
#if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX)
# error "HUF_TABLELOG_MAX is too large !"
#endif
/* ****************************************
* Static allocation
******************************************/
/* HUF buffer bounds */
#define HUF_CTABLEBOUND 129
#define HUF_BLOCKBOUND(size) (size + (size>>8) + 8) /* only true when incompressible is pre-filtered with fast heuristic */
#define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size)) /* Macro version, useful for static allocation */
/* static allocation of HUF's Compression Table */
/* this is a private definition, just exposed for allocation and strict aliasing purpose. never EVER access its members directly */
typedef size_t HUF_CElt; /* consider it an incomplete type */
#define HUF_CTABLE_SIZE_ST(maxSymbolValue) ((maxSymbolValue)+2) /* Use tables of size_t, for proper alignment */
#define HUF_CTABLE_SIZE(maxSymbolValue) (HUF_CTABLE_SIZE_ST(maxSymbolValue) * sizeof(size_t))
#define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \
HUF_CElt name[HUF_CTABLE_SIZE_ST(maxSymbolValue)] /* no final ; */
/* static allocation of HUF's DTable */
typedef U32 HUF_DTable;
#define HUF_DTABLE_SIZE(maxTableLog) (1 + (1<<(maxTableLog)))
#define HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog) \
HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) }
#define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \
HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) }
/* ****************************************
* Advanced decompression functions
******************************************/
/**
* Huffman flags bitset.
* For all flags, 0 is the default value.
*/
typedef enum {
/**
* If compiled with DYNAMIC_BMI2: Set flag only if the CPU supports BMI2 at runtime.
* Otherwise: Ignored.
*/
HUF_flags_bmi2 = (1 << 0),
/**
* If set: Test possible table depths to find the one that produces the smallest header + encoded size.
* If unset: Use heuristic to find the table depth.
*/
HUF_flags_optimalDepth = (1 << 1),
/**
* If set: If the previous table can encode the input, always reuse the previous table.
* If unset: If the previous table can encode the input, reuse the previous table if it results in a smaller output.
*/
HUF_flags_preferRepeat = (1 << 2),
/**
* If set: Sample the input and check if the sample is uncompressible, if it is then don't attempt to compress.
* If unset: Always histogram the entire input.
*/
HUF_flags_suspectUncompressible = (1 << 3),
/**
* If set: Don't use assembly implementations
* If unset: Allow using assembly implementations
*/
HUF_flags_disableAsm = (1 << 4),
/**
* If set: Don't use the fast decoding loop, always use the fallback decoding loop.
* If unset: Use the fast decoding loop when possible.
*/
HUF_flags_disableFast = (1 << 5)
} HUF_flags_e;
/* ****************************************
* HUF detailed API
* ****************************************/
#define HUF_OPTIMAL_DEPTH_THRESHOLD ZSTD_btultra
/*! HUF_compress() does the following:
* 1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")
* 2. (optional) refine tableLog using HUF_optimalTableLog()
* 3. build Huffman table from count using HUF_buildCTable()
* 4. save Huffman table to memory buffer using HUF_writeCTable()
* 5. encode the data stream using HUF_compress4X_usingCTable()
*
* The following API allows targeting specific sub-functions for advanced tasks.
* For example, it's possible to compress several blocks using the same 'CTable',
* or to save and regenerate 'CTable' using external methods.
*/
unsigned HUF_minTableLog(unsigned symbolCardinality);
unsigned HUF_cardinality(const unsigned* count, unsigned maxSymbolValue);
unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, void* workSpace,
size_t wkspSize, HUF_CElt* table, const unsigned* count, int flags); /* table is used as scratch space for building and testing tables, not a return value */
size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize);
size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
typedef enum {
HUF_repeat_none, /**< Cannot use the previous table */
HUF_repeat_check, /**< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1, 4}X_repeat */
HUF_repeat_valid /**< Can use the previous table and it is assumed to be valid */
} HUF_repeat;
/** HUF_compress4X_repeat() :
* Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
* If it uses hufTable it does not modify hufTable or repeat.
* If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
* If preferRepeat then the old table will always be used if valid.
* If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */
size_t HUF_compress4X_repeat(void* dst, size_t dstSize,
const void* src, size_t srcSize,
unsigned maxSymbolValue, unsigned tableLog,
void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
/** HUF_buildCTable_wksp() :
* Same as HUF_buildCTable(), but using externally allocated scratch buffer.
* `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.
*/
#define HUF_CTABLE_WORKSPACE_SIZE_U32 ((4 * (HUF_SYMBOLVALUE_MAX + 1)) + 192)
#define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
size_t HUF_buildCTable_wksp (HUF_CElt* tree,
const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,
void* workSpace, size_t wkspSize);
/*! HUF_readStats() :
* Read compact Huffman tree, saved by HUF_writeCTable().
* `huffWeight` is destination buffer.
* @return : size read from `src` , or an error Code .
* Note : Needed by HUF_readCTable() and HUF_readDTableXn() . */
size_t HUF_readStats(BYTE* huffWeight, size_t hwSize,
U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize);
/*! HUF_readStats_wksp() :
* Same as HUF_readStats() but takes an external workspace which must be
* 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE.
* If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
*/
#define HUF_READ_STATS_WORKSPACE_SIZE_U32 FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1)
#define HUF_READ_STATS_WORKSPACE_SIZE (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned))
size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize,
U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
const void* src, size_t srcSize,
void* workspace, size_t wkspSize,
int flags);
/** HUF_readCTable() :
* Loading a CTable saved with HUF_writeCTable() */
size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights);
/** HUF_getNbBitsFromCTable() :
* Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX
* Note 1 : is not inlined, as HUF_CElt definition is private */
U32 HUF_getNbBitsFromCTable(const HUF_CElt* symbolTable, U32 symbolValue);
/*
* HUF_decompress() does the following:
* 1. select the decompression algorithm (X1, X2) based on pre-computed heuristics
* 2. build Huffman table from save, using HUF_readDTableX?()
* 3. decode 1 or 4 segments in parallel using HUF_decompress?X?_usingDTable()
*/
/** HUF_selectDecoder() :
* Tells which decoder is likely to decode faster,
* based on a set of pre-computed metrics.
* @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 .
* Assumption : 0 < dstSize <= 128 KB */
U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);
/**
* The minimum workspace size for the `workSpace` used in
* HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp().
*
* The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when
* HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15.
* Buffer overflow errors may potentially occur if code modifications result in
* a required workspace size greater than that specified in the following
* macro.
*/
#define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9))
#define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))
/* ====================== */
/* single stream variants */
/* ====================== */
size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int flags);
/** HUF_compress1X_repeat() :
* Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
* If it uses hufTable it does not modify hufTable or repeat.
* If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
* If preferRepeat then the old table will always be used if valid.
* If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */
size_t HUF_compress1X_repeat(void* dst, size_t dstSize,
const void* src, size_t srcSize,
unsigned maxSymbolValue, unsigned tableLog,
void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
HUF_CElt* hufTable, HUF_repeat* repeat, int flags);
size_t HUF_decompress1X_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags); /**< double-symbols decoder */
#endif
/* BMI2 variants.
* If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
*/
size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
#endif
size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int flags);
size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int flags);
#ifndef HUF_FORCE_DECOMPRESS_X2
size_t HUF_readDTableX1_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
#endif
#ifndef HUF_FORCE_DECOMPRESS_X1
size_t HUF_readDTableX2_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int flags);
#endif
#endif /* HUF_H_298734234 */
#if defined (__cplusplus)
}
#endif

435
raytracer/nvpro_core/third_party/zstd/lib/common/mem.h vendored Normal file
View file

@ -0,0 +1,435 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef MEM_H_MODULE
#define MEM_H_MODULE
#if defined (__cplusplus)
extern "C" {
#endif
/*-****************************************
* Dependencies
******************************************/
#include <stddef.h> /* size_t, ptrdiff_t */
#include "compiler.h" /* __has_builtin */
#include "debug.h" /* DEBUG_STATIC_ASSERT */
#include "zstd_deps.h" /* ZSTD_memcpy */
/*-****************************************
* Compiler specifics
******************************************/
#if defined(_MSC_VER) /* Visual Studio */
# include <stdlib.h> /* _byteswap_ulong */
# include <intrin.h> /* _byteswap_* */
#endif
#if defined(__GNUC__)
# define MEM_STATIC static __inline __attribute__((unused))
#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
# define MEM_STATIC static inline
#elif defined(_MSC_VER)
# define MEM_STATIC static __inline
#else
# define MEM_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */
#endif
/*-**************************************************************
* Basic Types
*****************************************************************/
#if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
# if defined(_AIX)
# include <inttypes.h>
# else
# include <stdint.h> /* intptr_t */
# endif
typedef uint8_t BYTE;
typedef uint8_t U8;
typedef int8_t S8;
typedef uint16_t U16;
typedef int16_t S16;
typedef uint32_t U32;
typedef int32_t S32;
typedef uint64_t U64;
typedef int64_t S64;
#else
# include <limits.h>
#if CHAR_BIT != 8
# error "this implementation requires char to be exactly 8-bit type"
#endif
typedef unsigned char BYTE;
typedef unsigned char U8;
typedef signed char S8;
#if USHRT_MAX != 65535
# error "this implementation requires short to be exactly 16-bit type"
#endif
typedef unsigned short U16;
typedef signed short S16;
#if UINT_MAX != 4294967295
# error "this implementation requires int to be exactly 32-bit type"
#endif
typedef unsigned int U32;
typedef signed int S32;
/* note : there are no limits defined for long long type in C90.
* limits exist in C99, however, in such case, <stdint.h> is preferred */
typedef unsigned long long U64;
typedef signed long long S64;
#endif
/*-**************************************************************
* Memory I/O API
*****************************************************************/
/*=== Static platform detection ===*/
MEM_STATIC unsigned MEM_32bits(void);
MEM_STATIC unsigned MEM_64bits(void);
MEM_STATIC unsigned MEM_isLittleEndian(void);
/*=== Native unaligned read/write ===*/
MEM_STATIC U16 MEM_read16(const void* memPtr);
MEM_STATIC U32 MEM_read32(const void* memPtr);
MEM_STATIC U64 MEM_read64(const void* memPtr);
MEM_STATIC size_t MEM_readST(const void* memPtr);
MEM_STATIC void MEM_write16(void* memPtr, U16 value);
MEM_STATIC void MEM_write32(void* memPtr, U32 value);
MEM_STATIC void MEM_write64(void* memPtr, U64 value);
/*=== Little endian unaligned read/write ===*/
MEM_STATIC U16 MEM_readLE16(const void* memPtr);
MEM_STATIC U32 MEM_readLE24(const void* memPtr);
MEM_STATIC U32 MEM_readLE32(const void* memPtr);
MEM_STATIC U64 MEM_readLE64(const void* memPtr);
MEM_STATIC size_t MEM_readLEST(const void* memPtr);
MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val);
MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val);
MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32);
MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64);
MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val);
/*=== Big endian unaligned read/write ===*/
MEM_STATIC U32 MEM_readBE32(const void* memPtr);
MEM_STATIC U64 MEM_readBE64(const void* memPtr);
MEM_STATIC size_t MEM_readBEST(const void* memPtr);
MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32);
MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64);
MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val);
/*=== Byteswap ===*/
MEM_STATIC U32 MEM_swap32(U32 in);
MEM_STATIC U64 MEM_swap64(U64 in);
MEM_STATIC size_t MEM_swapST(size_t in);
/*-**************************************************************
* Memory I/O Implementation
*****************************************************************/
/* MEM_FORCE_MEMORY_ACCESS : For accessing unaligned memory:
* Method 0 : always use `memcpy()`. Safe and portable.
* Method 1 : Use compiler extension to set unaligned access.
* Method 2 : direct access. This method is portable but violate C standard.
* It can generate buggy code on targets depending on alignment.
* Default : method 1 if supported, else method 0
*/
#ifndef MEM_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */
# ifdef __GNUC__
# define MEM_FORCE_MEMORY_ACCESS 1
# endif
#endif
MEM_STATIC unsigned MEM_32bits(void) { return sizeof(size_t)==4; }
MEM_STATIC unsigned MEM_64bits(void) { return sizeof(size_t)==8; }
MEM_STATIC unsigned MEM_isLittleEndian(void)
{
#if defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
return 1;
#elif defined(__BYTE_ORDER__) && defined(__ORDER_BIG_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__)
return 0;
#elif defined(__clang__) && __LITTLE_ENDIAN__
return 1;
#elif defined(__clang__) && __BIG_ENDIAN__
return 0;
#elif defined(_MSC_VER) && (_M_AMD64 || _M_IX86)
return 1;
#elif defined(__DMC__) && defined(_M_IX86)
return 1;
#else
const union { U32 u; BYTE c[4]; } one = { 1 }; /* don't use static : performance detrimental */
return one.c[0];
#endif
}
#if defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==2)
/* violates C standard, by lying on structure alignment.
Only use if no other choice to achieve best performance on target platform */
MEM_STATIC U16 MEM_read16(const void* memPtr) { return *(const U16*) memPtr; }
MEM_STATIC U32 MEM_read32(const void* memPtr) { return *(const U32*) memPtr; }
MEM_STATIC U64 MEM_read64(const void* memPtr) { return *(const U64*) memPtr; }
MEM_STATIC size_t MEM_readST(const void* memPtr) { return *(const size_t*) memPtr; }
MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(U16*)memPtr = value; }
MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(U32*)memPtr = value; }
MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(U64*)memPtr = value; }
#elif defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==1)
typedef __attribute__((aligned(1))) U16 unalign16;
typedef __attribute__((aligned(1))) U32 unalign32;
typedef __attribute__((aligned(1))) U64 unalign64;
typedef __attribute__((aligned(1))) size_t unalignArch;
MEM_STATIC U16 MEM_read16(const void* ptr) { return *(const unalign16*)ptr; }
MEM_STATIC U32 MEM_read32(const void* ptr) { return *(const unalign32*)ptr; }
MEM_STATIC U64 MEM_read64(const void* ptr) { return *(const unalign64*)ptr; }
MEM_STATIC size_t MEM_readST(const void* ptr) { return *(const unalignArch*)ptr; }
MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(unalign16*)memPtr = value; }
MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(unalign32*)memPtr = value; }
MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(unalign64*)memPtr = value; }
#else
/* default method, safe and standard.
can sometimes prove slower */
MEM_STATIC U16 MEM_read16(const void* memPtr)
{
U16 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC U32 MEM_read32(const void* memPtr)
{
U32 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC U64 MEM_read64(const void* memPtr)
{
U64 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC size_t MEM_readST(const void* memPtr)
{
size_t val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val;
}
MEM_STATIC void MEM_write16(void* memPtr, U16 value)
{
ZSTD_memcpy(memPtr, &value, sizeof(value));
}
MEM_STATIC void MEM_write32(void* memPtr, U32 value)
{
ZSTD_memcpy(memPtr, &value, sizeof(value));
}
MEM_STATIC void MEM_write64(void* memPtr, U64 value)
{
ZSTD_memcpy(memPtr, &value, sizeof(value));
}
#endif /* MEM_FORCE_MEMORY_ACCESS */
MEM_STATIC U32 MEM_swap32_fallback(U32 in)
{
return ((in << 24) & 0xff000000 ) |
((in << 8) & 0x00ff0000 ) |
((in >> 8) & 0x0000ff00 ) |
((in >> 24) & 0x000000ff );
}
MEM_STATIC U32 MEM_swap32(U32 in)
{
#if defined(_MSC_VER) /* Visual Studio */
return _byteswap_ulong(in);
#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
|| (defined(__clang__) && __has_builtin(__builtin_bswap32))
return __builtin_bswap32(in);
#else
return MEM_swap32_fallback(in);
#endif
}
MEM_STATIC U64 MEM_swap64_fallback(U64 in)
{
return ((in << 56) & 0xff00000000000000ULL) |
((in << 40) & 0x00ff000000000000ULL) |
((in << 24) & 0x0000ff0000000000ULL) |
((in << 8) & 0x000000ff00000000ULL) |
((in >> 8) & 0x00000000ff000000ULL) |
((in >> 24) & 0x0000000000ff0000ULL) |
((in >> 40) & 0x000000000000ff00ULL) |
((in >> 56) & 0x00000000000000ffULL);
}
MEM_STATIC U64 MEM_swap64(U64 in)
{
#if defined(_MSC_VER) /* Visual Studio */
return _byteswap_uint64(in);
#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \
|| (defined(__clang__) && __has_builtin(__builtin_bswap64))
return __builtin_bswap64(in);
#else
return MEM_swap64_fallback(in);
#endif
}
MEM_STATIC size_t MEM_swapST(size_t in)
{
if (MEM_32bits())
return (size_t)MEM_swap32((U32)in);
else
return (size_t)MEM_swap64((U64)in);
}
/*=== Little endian r/w ===*/
MEM_STATIC U16 MEM_readLE16(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read16(memPtr);
else {
const BYTE* p = (const BYTE*)memPtr;
return (U16)(p[0] + (p[1]<<8));
}
}
MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val)
{
if (MEM_isLittleEndian()) {
MEM_write16(memPtr, val);
} else {
BYTE* p = (BYTE*)memPtr;
p[0] = (BYTE)val;
p[1] = (BYTE)(val>>8);
}
}
MEM_STATIC U32 MEM_readLE24(const void* memPtr)
{
return (U32)MEM_readLE16(memPtr) + ((U32)(((const BYTE*)memPtr)[2]) << 16);
}
MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val)
{
MEM_writeLE16(memPtr, (U16)val);
((BYTE*)memPtr)[2] = (BYTE)(val>>16);
}
MEM_STATIC U32 MEM_readLE32(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read32(memPtr);
else
return MEM_swap32(MEM_read32(memPtr));
}
MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32)
{
if (MEM_isLittleEndian())
MEM_write32(memPtr, val32);
else
MEM_write32(memPtr, MEM_swap32(val32));
}
MEM_STATIC U64 MEM_readLE64(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_read64(memPtr);
else
return MEM_swap64(MEM_read64(memPtr));
}
MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64)
{
if (MEM_isLittleEndian())
MEM_write64(memPtr, val64);
else
MEM_write64(memPtr, MEM_swap64(val64));
}
MEM_STATIC size_t MEM_readLEST(const void* memPtr)
{
if (MEM_32bits())
return (size_t)MEM_readLE32(memPtr);
else
return (size_t)MEM_readLE64(memPtr);
}
MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val)
{
if (MEM_32bits())
MEM_writeLE32(memPtr, (U32)val);
else
MEM_writeLE64(memPtr, (U64)val);
}
/*=== Big endian r/w ===*/
MEM_STATIC U32 MEM_readBE32(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_swap32(MEM_read32(memPtr));
else
return MEM_read32(memPtr);
}
MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32)
{
if (MEM_isLittleEndian())
MEM_write32(memPtr, MEM_swap32(val32));
else
MEM_write32(memPtr, val32);
}
MEM_STATIC U64 MEM_readBE64(const void* memPtr)
{
if (MEM_isLittleEndian())
return MEM_swap64(MEM_read64(memPtr));
else
return MEM_read64(memPtr);
}
MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64)
{
if (MEM_isLittleEndian())
MEM_write64(memPtr, MEM_swap64(val64));
else
MEM_write64(memPtr, val64);
}
MEM_STATIC size_t MEM_readBEST(const void* memPtr)
{
if (MEM_32bits())
return (size_t)MEM_readBE32(memPtr);
else
return (size_t)MEM_readBE64(memPtr);
}
MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val)
{
if (MEM_32bits())
MEM_writeBE32(memPtr, (U32)val);
else
MEM_writeBE64(memPtr, (U64)val);
}
/* code only tested on 32 and 64 bits systems */
MEM_STATIC void MEM_check(void) { DEBUG_STATIC_ASSERT((sizeof(size_t)==4) || (sizeof(size_t)==8)); }
#if defined (__cplusplus)
}
#endif
#endif /* MEM_H_MODULE */

371
raytracer/nvpro_core/third_party/zstd/lib/common/pool.c vendored Normal file
View file

@ -0,0 +1,371 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* ====== Dependencies ======= */
#include "../common/allocations.h" /* ZSTD_customCalloc, ZSTD_customFree */
#include "zstd_deps.h" /* size_t */
#include "debug.h" /* assert */
#include "pool.h"
/* ====== Compiler specifics ====== */
#if defined(_MSC_VER)
# pragma warning(disable : 4204) /* disable: C4204: non-constant aggregate initializer */
#endif
#ifdef ZSTD_MULTITHREAD
#include "threading.h" /* pthread adaptation */
/* A job is a function and an opaque argument */
typedef struct POOL_job_s {
POOL_function function;
void *opaque;
} POOL_job;
struct POOL_ctx_s {
ZSTD_customMem customMem;
/* Keep track of the threads */
ZSTD_pthread_t* threads;
size_t threadCapacity;
size_t threadLimit;
/* The queue is a circular buffer */
POOL_job *queue;
size_t queueHead;
size_t queueTail;
size_t queueSize;
/* The number of threads working on jobs */
size_t numThreadsBusy;
/* Indicates if the queue is empty */
int queueEmpty;
/* The mutex protects the queue */
ZSTD_pthread_mutex_t queueMutex;
/* Condition variable for pushers to wait on when the queue is full */
ZSTD_pthread_cond_t queuePushCond;
/* Condition variables for poppers to wait on when the queue is empty */
ZSTD_pthread_cond_t queuePopCond;
/* Indicates if the queue is shutting down */
int shutdown;
};
/* POOL_thread() :
* Work thread for the thread pool.
* Waits for jobs and executes them.
* @returns : NULL on failure else non-null.
*/
static void* POOL_thread(void* opaque) {
POOL_ctx* const ctx = (POOL_ctx*)opaque;
if (!ctx) { return NULL; }
for (;;) {
/* Lock the mutex and wait for a non-empty queue or until shutdown */
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
while ( ctx->queueEmpty
|| (ctx->numThreadsBusy >= ctx->threadLimit) ) {
if (ctx->shutdown) {
/* even if !queueEmpty, (possible if numThreadsBusy >= threadLimit),
* a few threads will be shutdown while !queueEmpty,
* but enough threads will remain active to finish the queue */
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
return opaque;
}
ZSTD_pthread_cond_wait(&ctx->queuePopCond, &ctx->queueMutex);
}
/* Pop a job off the queue */
{ POOL_job const job = ctx->queue[ctx->queueHead];
ctx->queueHead = (ctx->queueHead + 1) % ctx->queueSize;
ctx->numThreadsBusy++;
ctx->queueEmpty = (ctx->queueHead == ctx->queueTail);
/* Unlock the mutex, signal a pusher, and run the job */
ZSTD_pthread_cond_signal(&ctx->queuePushCond);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
job.function(job.opaque);
/* If the intended queue size was 0, signal after finishing job */
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
ctx->numThreadsBusy--;
ZSTD_pthread_cond_signal(&ctx->queuePushCond);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
}
} /* for (;;) */
assert(0); /* Unreachable */
}
/* ZSTD_createThreadPool() : public access point */
POOL_ctx* ZSTD_createThreadPool(size_t numThreads) {
return POOL_create (numThreads, 0);
}
POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
}
POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
ZSTD_customMem customMem)
{
POOL_ctx* ctx;
/* Check parameters */
if (!numThreads) { return NULL; }
/* Allocate the context and zero initialize */
ctx = (POOL_ctx*)ZSTD_customCalloc(sizeof(POOL_ctx), customMem);
if (!ctx) { return NULL; }
/* Initialize the job queue.
* It needs one extra space since one space is wasted to differentiate
* empty and full queues.
*/
ctx->queueSize = queueSize + 1;
ctx->queue = (POOL_job*)ZSTD_customCalloc(ctx->queueSize * sizeof(POOL_job), customMem);
ctx->queueHead = 0;
ctx->queueTail = 0;
ctx->numThreadsBusy = 0;
ctx->queueEmpty = 1;
{
int error = 0;
error |= ZSTD_pthread_mutex_init(&ctx->queueMutex, NULL);
error |= ZSTD_pthread_cond_init(&ctx->queuePushCond, NULL);
error |= ZSTD_pthread_cond_init(&ctx->queuePopCond, NULL);
if (error) { POOL_free(ctx); return NULL; }
}
ctx->shutdown = 0;
/* Allocate space for the thread handles */
ctx->threads = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), customMem);
ctx->threadCapacity = 0;
ctx->customMem = customMem;
/* Check for errors */
if (!ctx->threads || !ctx->queue) { POOL_free(ctx); return NULL; }
/* Initialize the threads */
{ size_t i;
for (i = 0; i < numThreads; ++i) {
if (ZSTD_pthread_create(&ctx->threads[i], NULL, &POOL_thread, ctx)) {
ctx->threadCapacity = i;
POOL_free(ctx);
return NULL;
} }
ctx->threadCapacity = numThreads;
ctx->threadLimit = numThreads;
}
return ctx;
}
/*! POOL_join() :
Shutdown the queue, wake any sleeping threads, and join all of the threads.
*/
static void POOL_join(POOL_ctx* ctx) {
/* Shut down the queue */
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
ctx->shutdown = 1;
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
/* Wake up sleeping threads */
ZSTD_pthread_cond_broadcast(&ctx->queuePushCond);
ZSTD_pthread_cond_broadcast(&ctx->queuePopCond);
/* Join all of the threads */
{ size_t i;
for (i = 0; i < ctx->threadCapacity; ++i) {
ZSTD_pthread_join(ctx->threads[i]); /* note : could fail */
} }
}
void POOL_free(POOL_ctx *ctx) {
if (!ctx) { return; }
POOL_join(ctx);
ZSTD_pthread_mutex_destroy(&ctx->queueMutex);
ZSTD_pthread_cond_destroy(&ctx->queuePushCond);
ZSTD_pthread_cond_destroy(&ctx->queuePopCond);
ZSTD_customFree(ctx->queue, ctx->customMem);
ZSTD_customFree(ctx->threads, ctx->customMem);
ZSTD_customFree(ctx, ctx->customMem);
}
/*! POOL_joinJobs() :
* Waits for all queued jobs to finish executing.
*/
void POOL_joinJobs(POOL_ctx* ctx) {
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
while(!ctx->queueEmpty || ctx->numThreadsBusy > 0) {
ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex);
}
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
}
void ZSTD_freeThreadPool (ZSTD_threadPool* pool) {
POOL_free (pool);
}
size_t POOL_sizeof(const POOL_ctx* ctx) {
if (ctx==NULL) return 0; /* supports sizeof NULL */
return sizeof(*ctx)
+ ctx->queueSize * sizeof(POOL_job)
+ ctx->threadCapacity * sizeof(ZSTD_pthread_t);
}
/* @return : 0 on success, 1 on error */
static int POOL_resize_internal(POOL_ctx* ctx, size_t numThreads)
{
if (numThreads <= ctx->threadCapacity) {
if (!numThreads) return 1;
ctx->threadLimit = numThreads;
return 0;
}
/* numThreads > threadCapacity */
{ ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customCalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem);
if (!threadPool) return 1;
/* replace existing thread pool */
ZSTD_memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool));
ZSTD_customFree(ctx->threads, ctx->customMem);
ctx->threads = threadPool;
/* Initialize additional threads */
{ size_t threadId;
for (threadId = ctx->threadCapacity; threadId < numThreads; ++threadId) {
if (ZSTD_pthread_create(&threadPool[threadId], NULL, &POOL_thread, ctx)) {
ctx->threadCapacity = threadId;
return 1;
} }
} }
/* successfully expanded */
ctx->threadCapacity = numThreads;
ctx->threadLimit = numThreads;
return 0;
}
/* @return : 0 on success, 1 on error */
int POOL_resize(POOL_ctx* ctx, size_t numThreads)
{
int result;
if (ctx==NULL) return 1;
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
result = POOL_resize_internal(ctx, numThreads);
ZSTD_pthread_cond_broadcast(&ctx->queuePopCond);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
return result;
}
/**
* Returns 1 if the queue is full and 0 otherwise.
*
* When queueSize is 1 (pool was created with an intended queueSize of 0),
* then a queue is empty if there is a thread free _and_ no job is waiting.
*/
static int isQueueFull(POOL_ctx const* ctx) {
if (ctx->queueSize > 1) {
return ctx->queueHead == ((ctx->queueTail + 1) % ctx->queueSize);
} else {
return (ctx->numThreadsBusy == ctx->threadLimit) ||
!ctx->queueEmpty;
}
}
static void
POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
{
POOL_job job;
job.function = function;
job.opaque = opaque;
assert(ctx != NULL);
if (ctx->shutdown) return;
ctx->queueEmpty = 0;
ctx->queue[ctx->queueTail] = job;
ctx->queueTail = (ctx->queueTail + 1) % ctx->queueSize;
ZSTD_pthread_cond_signal(&ctx->queuePopCond);
}
void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque)
{
assert(ctx != NULL);
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
/* Wait until there is space in the queue for the new job */
while (isQueueFull(ctx) && (!ctx->shutdown)) {
ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex);
}
POOL_add_internal(ctx, function, opaque);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
}
int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque)
{
assert(ctx != NULL);
ZSTD_pthread_mutex_lock(&ctx->queueMutex);
if (isQueueFull(ctx)) {
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
return 0;
}
POOL_add_internal(ctx, function, opaque);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
return 1;
}
#else /* ZSTD_MULTITHREAD not defined */
/* ========================== */
/* No multi-threading support */
/* ========================== */
/* We don't need any data, but if it is empty, malloc() might return NULL. */
struct POOL_ctx_s {
int dummy;
};
static POOL_ctx g_poolCtx;
POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
}
POOL_ctx*
POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem)
{
(void)numThreads;
(void)queueSize;
(void)customMem;
return &g_poolCtx;
}
void POOL_free(POOL_ctx* ctx) {
assert(!ctx || ctx == &g_poolCtx);
(void)ctx;
}
void POOL_joinJobs(POOL_ctx* ctx){
assert(!ctx || ctx == &g_poolCtx);
(void)ctx;
}
int POOL_resize(POOL_ctx* ctx, size_t numThreads) {
(void)ctx; (void)numThreads;
return 0;
}
void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque) {
(void)ctx;
function(opaque);
}
int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) {
(void)ctx;
function(opaque);
return 1;
}
size_t POOL_sizeof(const POOL_ctx* ctx) {
if (ctx==NULL) return 0; /* supports sizeof NULL */
assert(ctx == &g_poolCtx);
return sizeof(*ctx);
}
#endif /* ZSTD_MULTITHREAD */

90
raytracer/nvpro_core/third_party/zstd/lib/common/pool.h vendored Normal file
View file

@ -0,0 +1,90 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef POOL_H
#define POOL_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "zstd_deps.h"
#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_customMem */
#include "../zstd.h"
typedef struct POOL_ctx_s POOL_ctx;
/*! POOL_create() :
* Create a thread pool with at most `numThreads` threads.
* `numThreads` must be at least 1.
* The maximum number of queued jobs before blocking is `queueSize`.
* @return : POOL_ctx pointer on success, else NULL.
*/
POOL_ctx* POOL_create(size_t numThreads, size_t queueSize);
POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
ZSTD_customMem customMem);
/*! POOL_free() :
* Free a thread pool returned by POOL_create().
*/
void POOL_free(POOL_ctx* ctx);
/*! POOL_joinJobs() :
* Waits for all queued jobs to finish executing.
*/
void POOL_joinJobs(POOL_ctx* ctx);
/*! POOL_resize() :
* Expands or shrinks pool's number of threads.
* This is more efficient than releasing + creating a new context,
* since it tries to preserve and re-use existing threads.
* `numThreads` must be at least 1.
* @return : 0 when resize was successful,
* !0 (typically 1) if there is an error.
* note : only numThreads can be resized, queueSize remains unchanged.
*/
int POOL_resize(POOL_ctx* ctx, size_t numThreads);
/*! POOL_sizeof() :
* @return threadpool memory usage
* note : compatible with NULL (returns 0 in this case)
*/
size_t POOL_sizeof(const POOL_ctx* ctx);
/*! POOL_function :
* The function type that can be added to a thread pool.
*/
typedef void (*POOL_function)(void*);
/*! POOL_add() :
* Add the job `function(opaque)` to the thread pool. `ctx` must be valid.
* Possibly blocks until there is room in the queue.
* Note : The function may be executed asynchronously,
* therefore, `opaque` must live until function has been completed.
*/
void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque);
/*! POOL_tryAdd() :
* Add the job `function(opaque)` to thread pool _if_ a queue slot is available.
* Returns immediately even if not (does not block).
* @return : 1 if successful, 0 if not.
*/
int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque);
#if defined (__cplusplus)
}
#endif
#endif

156
raytracer/nvpro_core/third_party/zstd/lib/common/portability_macros.h vendored Normal file
View file

@ -0,0 +1,156 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_PORTABILITY_MACROS_H
#define ZSTD_PORTABILITY_MACROS_H
/**
* This header file contains macro definitions to support portability.
* This header is shared between C and ASM code, so it MUST only
* contain macro definitions. It MUST not contain any C code.
*
* This header ONLY defines macros to detect platforms/feature support.
*
*/
/* compat. with non-clang compilers */
#ifndef __has_attribute
#define __has_attribute(x) 0
#endif
/* compat. with non-clang compilers */
#ifndef __has_builtin
# define __has_builtin(x) 0
#endif
/* compat. with non-clang compilers */
#ifndef __has_feature
# define __has_feature(x) 0
#endif
/* detects whether we are being compiled under msan */
#ifndef ZSTD_MEMORY_SANITIZER
# if __has_feature(memory_sanitizer)
# define ZSTD_MEMORY_SANITIZER 1
# else
# define ZSTD_MEMORY_SANITIZER 0
# endif
#endif
/* detects whether we are being compiled under asan */
#ifndef ZSTD_ADDRESS_SANITIZER
# if __has_feature(address_sanitizer)
# define ZSTD_ADDRESS_SANITIZER 1
# elif defined(__SANITIZE_ADDRESS__)
# define ZSTD_ADDRESS_SANITIZER 1
# else
# define ZSTD_ADDRESS_SANITIZER 0
# endif
#endif
/* detects whether we are being compiled under dfsan */
#ifndef ZSTD_DATAFLOW_SANITIZER
# if __has_feature(dataflow_sanitizer)
# define ZSTD_DATAFLOW_SANITIZER 1
# else
# define ZSTD_DATAFLOW_SANITIZER 0
# endif
#endif
/* Mark the internal assembly functions as hidden */
#ifdef __ELF__
# define ZSTD_HIDE_ASM_FUNCTION(func) .hidden func
#else
# define ZSTD_HIDE_ASM_FUNCTION(func)
#endif
/* Enable runtime BMI2 dispatch based on the CPU.
* Enabled for clang & gcc >=4.8 on x86 when BMI2 isn't enabled by default.
*/
#ifndef DYNAMIC_BMI2
#if ((defined(__clang__) && __has_attribute(__target__)) \
|| (defined(__GNUC__) \
&& (__GNUC__ >= 5 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8)))) \
&& (defined(__x86_64__) || defined(_M_X64)) \
&& !defined(__BMI2__)
# define DYNAMIC_BMI2 1
#else
# define DYNAMIC_BMI2 0
#endif
#endif
/**
* Only enable assembly for GNUC compatible compilers,
* because other platforms may not support GAS assembly syntax.
*
* Only enable assembly for Linux / MacOS, other platforms may
* work, but they haven't been tested. This could likely be
* extended to BSD systems.
*
* Disable assembly when MSAN is enabled, because MSAN requires
* 100% of code to be instrumented to work.
*/
#if defined(__GNUC__)
# if defined(__linux__) || defined(__linux) || defined(__APPLE__)
# if ZSTD_MEMORY_SANITIZER
# define ZSTD_ASM_SUPPORTED 0
# elif ZSTD_DATAFLOW_SANITIZER
# define ZSTD_ASM_SUPPORTED 0
# else
# define ZSTD_ASM_SUPPORTED 1
# endif
# else
# define ZSTD_ASM_SUPPORTED 0
# endif
#else
# define ZSTD_ASM_SUPPORTED 0
#endif
/**
* Determines whether we should enable assembly for x86-64
* with BMI2.
*
* Enable if all of the following conditions hold:
* - ASM hasn't been explicitly disabled by defining ZSTD_DISABLE_ASM
* - Assembly is supported
* - We are compiling for x86-64 and either:
* - DYNAMIC_BMI2 is enabled
* - BMI2 is supported at compile time
*/
#if !defined(ZSTD_DISABLE_ASM) && \
ZSTD_ASM_SUPPORTED && \
defined(__x86_64__) && \
(DYNAMIC_BMI2 || defined(__BMI2__))
# define ZSTD_ENABLE_ASM_X86_64_BMI2 1
#else
# define ZSTD_ENABLE_ASM_X86_64_BMI2 0
#endif
/*
* For x86 ELF targets, add .note.gnu.property section for Intel CET in
* assembly sources when CET is enabled.
*
* Additionally, any function that may be called indirectly must begin
* with ZSTD_CET_ENDBRANCH.
*/
#if defined(__ELF__) && (defined(__x86_64__) || defined(__i386__)) \
&& defined(__has_include)
# if __has_include(<cet.h>)
# include <cet.h>
# define ZSTD_CET_ENDBRANCH _CET_ENDBR
# endif
#endif
#ifndef ZSTD_CET_ENDBRANCH
# define ZSTD_CET_ENDBRANCH
#endif
#endif /* ZSTD_PORTABILITY_MACROS_H */

176
raytracer/nvpro_core/third_party/zstd/lib/common/threading.c vendored Normal file
View file

@ -0,0 +1,176 @@
/**
* Copyright (c) 2016 Tino Reichardt
* All rights reserved.
*
* You can contact the author at:
* - zstdmt source repository: https://github.com/mcmilk/zstdmt
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/**
* This file will hold wrapper for systems, which do not support pthreads
*/
#include "threading.h"
/* create fake symbol to avoid empty translation unit warning */
int g_ZSTD_threading_useless_symbol;
#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
/**
* Windows minimalist Pthread Wrapper
*/
/* === Dependencies === */
#include <process.h>
#include <errno.h>
/* === Implementation === */
typedef struct {
void* (*start_routine)(void*);
void* arg;
int initialized;
ZSTD_pthread_cond_t initialized_cond;
ZSTD_pthread_mutex_t initialized_mutex;
} ZSTD_thread_params_t;
static unsigned __stdcall worker(void *arg)
{
void* (*start_routine)(void*);
void* thread_arg;
/* Initialized thread_arg and start_routine and signal main thread that we don't need it
* to wait any longer.
*/
{
ZSTD_thread_params_t* thread_param = (ZSTD_thread_params_t*)arg;
thread_arg = thread_param->arg;
start_routine = thread_param->start_routine;
/* Signal main thread that we are running and do not depend on its memory anymore */
ZSTD_pthread_mutex_lock(&thread_param->initialized_mutex);
thread_param->initialized = 1;
ZSTD_pthread_cond_signal(&thread_param->initialized_cond);
ZSTD_pthread_mutex_unlock(&thread_param->initialized_mutex);
}
start_routine(thread_arg);
return 0;
}
int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
void* (*start_routine) (void*), void* arg)
{
ZSTD_thread_params_t thread_param;
(void)unused;
thread_param.start_routine = start_routine;
thread_param.arg = arg;
thread_param.initialized = 0;
*thread = NULL;
/* Setup thread initialization synchronization */
if(ZSTD_pthread_cond_init(&thread_param.initialized_cond, NULL)) {
/* Should never happen on Windows */
return -1;
}
if(ZSTD_pthread_mutex_init(&thread_param.initialized_mutex, NULL)) {
/* Should never happen on Windows */
ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
return -1;
}
/* Spawn thread */
*thread = (HANDLE)_beginthreadex(NULL, 0, worker, &thread_param, 0, NULL);
if (!thread) {
ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
return errno;
}
/* Wait for thread to be initialized */
ZSTD_pthread_mutex_lock(&thread_param.initialized_mutex);
while(!thread_param.initialized) {
ZSTD_pthread_cond_wait(&thread_param.initialized_cond, &thread_param.initialized_mutex);
}
ZSTD_pthread_mutex_unlock(&thread_param.initialized_mutex);
ZSTD_pthread_mutex_destroy(&thread_param.initialized_mutex);
ZSTD_pthread_cond_destroy(&thread_param.initialized_cond);
return 0;
}
int ZSTD_pthread_join(ZSTD_pthread_t thread)
{
DWORD result;
if (!thread) return 0;
result = WaitForSingleObject(thread, INFINITE);
CloseHandle(thread);
switch (result) {
case WAIT_OBJECT_0:
return 0;
case WAIT_ABANDONED:
return EINVAL;
default:
return GetLastError();
}
}
#endif /* ZSTD_MULTITHREAD */
#if defined(ZSTD_MULTITHREAD) && DEBUGLEVEL >= 1 && !defined(_WIN32)
#define ZSTD_DEPS_NEED_MALLOC
#include "zstd_deps.h"
int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr)
{
*mutex = (pthread_mutex_t*)ZSTD_malloc(sizeof(pthread_mutex_t));
if (!*mutex)
return 1;
return pthread_mutex_init(*mutex, attr);
}
int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex)
{
if (!*mutex)
return 0;
{
int const ret = pthread_mutex_destroy(*mutex);
ZSTD_free(*mutex);
return ret;
}
}
int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr)
{
*cond = (pthread_cond_t*)ZSTD_malloc(sizeof(pthread_cond_t));
if (!*cond)
return 1;
return pthread_cond_init(*cond, attr);
}
int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond)
{
if (!*cond)
return 0;
{
int const ret = pthread_cond_destroy(*cond);
ZSTD_free(*cond);
return ret;
}
}
#endif

150
raytracer/nvpro_core/third_party/zstd/lib/common/threading.h vendored Normal file
View file

@ -0,0 +1,150 @@
/**
* Copyright (c) 2016 Tino Reichardt
* All rights reserved.
*
* You can contact the author at:
* - zstdmt source repository: https://github.com/mcmilk/zstdmt
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef THREADING_H_938743
#define THREADING_H_938743
#include "debug.h"
#if defined (__cplusplus)
extern "C" {
#endif
#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
/**
* Windows minimalist Pthread Wrapper
*/
#ifdef WINVER
# undef WINVER
#endif
#define WINVER 0x0600
#ifdef _WIN32_WINNT
# undef _WIN32_WINNT
#endif
#define _WIN32_WINNT 0x0600
#ifndef WIN32_LEAN_AND_MEAN
# define WIN32_LEAN_AND_MEAN
#endif
#undef ERROR /* reported already defined on VS 2015 (Rich Geldreich) */
#include <windows.h>
#undef ERROR
#define ERROR(name) ZSTD_ERROR(name)
/* mutex */
#define ZSTD_pthread_mutex_t CRITICAL_SECTION
#define ZSTD_pthread_mutex_init(a, b) ((void)(b), InitializeCriticalSection((a)), 0)
#define ZSTD_pthread_mutex_destroy(a) DeleteCriticalSection((a))
#define ZSTD_pthread_mutex_lock(a) EnterCriticalSection((a))
#define ZSTD_pthread_mutex_unlock(a) LeaveCriticalSection((a))
/* condition variable */
#define ZSTD_pthread_cond_t CONDITION_VARIABLE
#define ZSTD_pthread_cond_init(a, b) ((void)(b), InitializeConditionVariable((a)), 0)
#define ZSTD_pthread_cond_destroy(a) ((void)(a))
#define ZSTD_pthread_cond_wait(a, b) SleepConditionVariableCS((a), (b), INFINITE)
#define ZSTD_pthread_cond_signal(a) WakeConditionVariable((a))
#define ZSTD_pthread_cond_broadcast(a) WakeAllConditionVariable((a))
/* ZSTD_pthread_create() and ZSTD_pthread_join() */
typedef HANDLE ZSTD_pthread_t;
int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused,
void* (*start_routine) (void*), void* arg);
int ZSTD_pthread_join(ZSTD_pthread_t thread);
/**
* add here more wrappers as required
*/
#elif defined(ZSTD_MULTITHREAD) /* posix assumed ; need a better detection method */
/* === POSIX Systems === */
# include <pthread.h>
#if DEBUGLEVEL < 1
#define ZSTD_pthread_mutex_t pthread_mutex_t
#define ZSTD_pthread_mutex_init(a, b) pthread_mutex_init((a), (b))
#define ZSTD_pthread_mutex_destroy(a) pthread_mutex_destroy((a))
#define ZSTD_pthread_mutex_lock(a) pthread_mutex_lock((a))
#define ZSTD_pthread_mutex_unlock(a) pthread_mutex_unlock((a))
#define ZSTD_pthread_cond_t pthread_cond_t
#define ZSTD_pthread_cond_init(a, b) pthread_cond_init((a), (b))
#define ZSTD_pthread_cond_destroy(a) pthread_cond_destroy((a))
#define ZSTD_pthread_cond_wait(a, b) pthread_cond_wait((a), (b))
#define ZSTD_pthread_cond_signal(a) pthread_cond_signal((a))
#define ZSTD_pthread_cond_broadcast(a) pthread_cond_broadcast((a))
#define ZSTD_pthread_t pthread_t
#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
#define ZSTD_pthread_join(a) pthread_join((a),NULL)
#else /* DEBUGLEVEL >= 1 */
/* Debug implementation of threading.
* In this implementation we use pointers for mutexes and condition variables.
* This way, if we forget to init/destroy them the program will crash or ASAN
* will report leaks.
*/
#define ZSTD_pthread_mutex_t pthread_mutex_t*
int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr);
int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex);
#define ZSTD_pthread_mutex_lock(a) pthread_mutex_lock(*(a))
#define ZSTD_pthread_mutex_unlock(a) pthread_mutex_unlock(*(a))
#define ZSTD_pthread_cond_t pthread_cond_t*
int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr);
int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond);
#define ZSTD_pthread_cond_wait(a, b) pthread_cond_wait(*(a), *(b))
#define ZSTD_pthread_cond_signal(a) pthread_cond_signal(*(a))
#define ZSTD_pthread_cond_broadcast(a) pthread_cond_broadcast(*(a))
#define ZSTD_pthread_t pthread_t
#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d))
#define ZSTD_pthread_join(a) pthread_join((a),NULL)
#endif
#else /* ZSTD_MULTITHREAD not defined */
/* No multithreading support */
typedef int ZSTD_pthread_mutex_t;
#define ZSTD_pthread_mutex_init(a, b) ((void)(a), (void)(b), 0)
#define ZSTD_pthread_mutex_destroy(a) ((void)(a))
#define ZSTD_pthread_mutex_lock(a) ((void)(a))
#define ZSTD_pthread_mutex_unlock(a) ((void)(a))
typedef int ZSTD_pthread_cond_t;
#define ZSTD_pthread_cond_init(a, b) ((void)(a), (void)(b), 0)
#define ZSTD_pthread_cond_destroy(a) ((void)(a))
#define ZSTD_pthread_cond_wait(a, b) ((void)(a), (void)(b))
#define ZSTD_pthread_cond_signal(a) ((void)(a))
#define ZSTD_pthread_cond_broadcast(a) ((void)(a))
/* do not use ZSTD_pthread_t */
#endif /* ZSTD_MULTITHREAD */
#if defined (__cplusplus)
}
#endif
#endif /* THREADING_H_938743 */

24
raytracer/nvpro_core/third_party/zstd/lib/common/xxhash.c vendored Normal file
View file

@ -0,0 +1,24 @@
/*
* xxHash - Fast Hash algorithm
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - xxHash homepage: https://cyan4973.github.io/xxHash/
* - xxHash source repository : https://github.com/Cyan4973/xxHash
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*
* xxhash.c instantiates functions defined in xxhash.h
*/
#define XXH_STATIC_LINKING_ONLY /* access advanced declarations */
#define XXH_IMPLEMENTATION /* access definitions */
#include "xxhash.h"

5686
raytracer/nvpro_core/third_party/zstd/lib/common/xxhash.h vendored Normal file
View file

File diff suppressed because it is too large Load diff

48
raytracer/nvpro_core/third_party/zstd/lib/common/zstd_common.c vendored Normal file
View file

@ -0,0 +1,48 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*-*************************************
* Dependencies
***************************************/
#define ZSTD_DEPS_NEED_MALLOC
#include "error_private.h"
#include "zstd_internal.h"
/*-****************************************
* Version
******************************************/
unsigned ZSTD_versionNumber(void) { return ZSTD_VERSION_NUMBER; }
const char* ZSTD_versionString(void) { return ZSTD_VERSION_STRING; }
/*-****************************************
* ZSTD Error Management
******************************************/
#undef ZSTD_isError /* defined within zstd_internal.h */
/*! ZSTD_isError() :
* tells if a return value is an error code
* symbol is required for external callers */
unsigned ZSTD_isError(size_t code) { return ERR_isError(code); }
/*! ZSTD_getErrorName() :
* provides error code string from function result (useful for debugging) */
const char* ZSTD_getErrorName(size_t code) { return ERR_getErrorName(code); }
/*! ZSTD_getError() :
* convert a `size_t` function result into a proper ZSTD_errorCode enum */
ZSTD_ErrorCode ZSTD_getErrorCode(size_t code) { return ERR_getErrorCode(code); }
/*! ZSTD_getErrorString() :
* provides error code string from enum */
const char* ZSTD_getErrorString(ZSTD_ErrorCode code) { return ERR_getErrorString(code); }

111
raytracer/nvpro_core/third_party/zstd/lib/common/zstd_deps.h vendored Normal file
View file

@ -0,0 +1,111 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* This file provides common libc dependencies that zstd requires.
* The purpose is to allow replacing this file with a custom implementation
* to compile zstd without libc support.
*/
/* Need:
* NULL
* INT_MAX
* UINT_MAX
* ZSTD_memcpy()
* ZSTD_memset()
* ZSTD_memmove()
*/
#ifndef ZSTD_DEPS_COMMON
#define ZSTD_DEPS_COMMON
#include <limits.h>
#include <stddef.h>
#include <string.h>
#if defined(__GNUC__) && __GNUC__ >= 4
# define ZSTD_memcpy(d,s,l) __builtin_memcpy((d),(s),(l))
# define ZSTD_memmove(d,s,l) __builtin_memmove((d),(s),(l))
# define ZSTD_memset(p,v,l) __builtin_memset((p),(v),(l))
#else
# define ZSTD_memcpy(d,s,l) memcpy((d),(s),(l))
# define ZSTD_memmove(d,s,l) memmove((d),(s),(l))
# define ZSTD_memset(p,v,l) memset((p),(v),(l))
#endif
#endif /* ZSTD_DEPS_COMMON */
/* Need:
* ZSTD_malloc()
* ZSTD_free()
* ZSTD_calloc()
*/
#ifdef ZSTD_DEPS_NEED_MALLOC
#ifndef ZSTD_DEPS_MALLOC
#define ZSTD_DEPS_MALLOC
#include <stdlib.h>
#define ZSTD_malloc(s) malloc(s)
#define ZSTD_calloc(n,s) calloc((n), (s))
#define ZSTD_free(p) free((p))
#endif /* ZSTD_DEPS_MALLOC */
#endif /* ZSTD_DEPS_NEED_MALLOC */
/*
* Provides 64-bit math support.
* Need:
* U64 ZSTD_div64(U64 dividend, U32 divisor)
*/
#ifdef ZSTD_DEPS_NEED_MATH64
#ifndef ZSTD_DEPS_MATH64
#define ZSTD_DEPS_MATH64
#define ZSTD_div64(dividend, divisor) ((dividend) / (divisor))
#endif /* ZSTD_DEPS_MATH64 */
#endif /* ZSTD_DEPS_NEED_MATH64 */
/* Need:
* assert()
*/
#ifdef ZSTD_DEPS_NEED_ASSERT
#ifndef ZSTD_DEPS_ASSERT
#define ZSTD_DEPS_ASSERT
#include <assert.h>
#endif /* ZSTD_DEPS_ASSERT */
#endif /* ZSTD_DEPS_NEED_ASSERT */
/* Need:
* ZSTD_DEBUG_PRINT()
*/
#ifdef ZSTD_DEPS_NEED_IO
#ifndef ZSTD_DEPS_IO
#define ZSTD_DEPS_IO
#include <stdio.h>
#define ZSTD_DEBUG_PRINT(...) fprintf(stderr, __VA_ARGS__)
#endif /* ZSTD_DEPS_IO */
#endif /* ZSTD_DEPS_NEED_IO */
/* Only requested when <stdint.h> is known to be present.
* Need:
* intptr_t
*/
#ifdef ZSTD_DEPS_NEED_STDINT
#ifndef ZSTD_DEPS_STDINT
#define ZSTD_DEPS_STDINT
#include <stdint.h>
#endif /* ZSTD_DEPS_STDINT */
#endif /* ZSTD_DEPS_NEED_STDINT */

392
raytracer/nvpro_core/third_party/zstd/lib/common/zstd_internal.h vendored Normal file
View file

@ -0,0 +1,392 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_CCOMMON_H_MODULE
#define ZSTD_CCOMMON_H_MODULE
/* this module contains definitions which must be identical
* across compression, decompression and dictBuilder.
* It also contains a few functions useful to at least 2 of them
* and which benefit from being inlined */
/*-*************************************
* Dependencies
***************************************/
#include "compiler.h"
#include "cpu.h"
#include "mem.h"
#include "debug.h" /* assert, DEBUGLOG, RAWLOG, g_debuglevel */
#include "error_private.h"
#define ZSTD_STATIC_LINKING_ONLY
#include "../zstd.h"
#define FSE_STATIC_LINKING_ONLY
#include "fse.h"
#include "huf.h"
#ifndef XXH_STATIC_LINKING_ONLY
# define XXH_STATIC_LINKING_ONLY /* XXH64_state_t */
#endif
#include "xxhash.h" /* XXH_reset, update, digest */
#ifndef ZSTD_NO_TRACE
# include "zstd_trace.h"
#else
# define ZSTD_TRACE 0
#endif
#if defined (__cplusplus)
extern "C" {
#endif
/* ---- static assert (debug) --- */
#define ZSTD_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c)
#define ZSTD_isError ERR_isError /* for inlining */
#define FSE_isError ERR_isError
#define HUF_isError ERR_isError
/*-*************************************
* shared macros
***************************************/
#undef MIN
#undef MAX
#define MIN(a,b) ((a)<(b) ? (a) : (b))
#define MAX(a,b) ((a)>(b) ? (a) : (b))
#define BOUNDED(min,val,max) (MAX(min,MIN(val,max)))
/*-*************************************
* Common constants
***************************************/
#define ZSTD_OPT_NUM (1<<12)
#define ZSTD_REP_NUM 3 /* number of repcodes */
static UNUSED_ATTR const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 };
#define KB *(1 <<10)
#define MB *(1 <<20)
#define GB *(1U<<30)
#define BIT7 128
#define BIT6 64
#define BIT5 32
#define BIT4 16
#define BIT1 2
#define BIT0 1
#define ZSTD_WINDOWLOG_ABSOLUTEMIN 10
static UNUSED_ATTR const size_t ZSTD_fcs_fieldSize[4] = { 0, 2, 4, 8 };
static UNUSED_ATTR const size_t ZSTD_did_fieldSize[4] = { 0, 1, 2, 4 };
#define ZSTD_FRAMEIDSIZE 4 /* magic number size */
#define ZSTD_BLOCKHEADERSIZE 3 /* C standard doesn't allow `static const` variable to be init using another `static const` variable */
static UNUSED_ATTR const size_t ZSTD_blockHeaderSize = ZSTD_BLOCKHEADERSIZE;
typedef enum { bt_raw, bt_rle, bt_compressed, bt_reserved } blockType_e;
#define ZSTD_FRAMECHECKSUMSIZE 4
#define MIN_SEQUENCES_SIZE 1 /* nbSeq==0 */
#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */) /* for a non-null block */
#define MIN_LITERALS_FOR_4_STREAMS 6
typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingType_e;
#define LONGNBSEQ 0x7F00
#define MINMATCH 3
#define Litbits 8
#define LitHufLog 11
#define MaxLit ((1<<Litbits) - 1)
#define MaxML 52
#define MaxLL 35
#define DefaultMaxOff 28
#define MaxOff 31
#define MaxSeq MAX(MaxLL, MaxML) /* Assumption : MaxOff < MaxLL,MaxML */
#define MLFSELog 9
#define LLFSELog 9
#define OffFSELog 8
#define MaxFSELog MAX(MAX(MLFSELog, LLFSELog), OffFSELog)
#define MaxMLBits 16
#define MaxLLBits 16
#define ZSTD_MAX_HUF_HEADER_SIZE 128 /* header + <= 127 byte tree description */
/* Each table cannot take more than #symbols * FSELog bits */
#define ZSTD_MAX_FSE_HEADERS_SIZE (((MaxML + 1) * MLFSELog + (MaxLL + 1) * LLFSELog + (MaxOff + 1) * OffFSELog + 7) / 8)
static UNUSED_ATTR const U8 LL_bits[MaxLL+1] = {
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
1, 1, 1, 1, 2, 2, 3, 3,
4, 6, 7, 8, 9,10,11,12,
13,14,15,16
};
static UNUSED_ATTR const S16 LL_defaultNorm[MaxLL+1] = {
4, 3, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 1, 1, 1,
2, 2, 2, 2, 2, 2, 2, 2,
2, 3, 2, 1, 1, 1, 1, 1,
-1,-1,-1,-1
};
#define LL_DEFAULTNORMLOG 6 /* for static allocation */
static UNUSED_ATTR const U32 LL_defaultNormLog = LL_DEFAULTNORMLOG;
static UNUSED_ATTR const U8 ML_bits[MaxML+1] = {
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
1, 1, 1, 1, 2, 2, 3, 3,
4, 4, 5, 7, 8, 9,10,11,
12,13,14,15,16
};
static UNUSED_ATTR const S16 ML_defaultNorm[MaxML+1] = {
1, 4, 3, 2, 2, 2, 2, 2,
2, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1,-1,-1,
-1,-1,-1,-1,-1
};
#define ML_DEFAULTNORMLOG 6 /* for static allocation */
static UNUSED_ATTR const U32 ML_defaultNormLog = ML_DEFAULTNORMLOG;
static UNUSED_ATTR const S16 OF_defaultNorm[DefaultMaxOff+1] = {
1, 1, 1, 1, 1, 1, 2, 2,
2, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
-1,-1,-1,-1,-1
};
#define OF_DEFAULTNORMLOG 5 /* for static allocation */
static UNUSED_ATTR const U32 OF_defaultNormLog = OF_DEFAULTNORMLOG;
/*-*******************************************
* Shared functions to include for inlining
*********************************************/
static void ZSTD_copy8(void* dst, const void* src) {
#if defined(ZSTD_ARCH_ARM_NEON)
vst1_u8((uint8_t*)dst, vld1_u8((const uint8_t*)src));
#else
ZSTD_memcpy(dst, src, 8);
#endif
}
#define COPY8(d,s) { ZSTD_copy8(d,s); d+=8; s+=8; }
/* Need to use memmove here since the literal buffer can now be located within
the dst buffer. In circumstances where the op "catches up" to where the
literal buffer is, there can be partial overlaps in this call on the final
copy if the literal is being shifted by less than 16 bytes. */
static void ZSTD_copy16(void* dst, const void* src) {
#if defined(ZSTD_ARCH_ARM_NEON)
vst1q_u8((uint8_t*)dst, vld1q_u8((const uint8_t*)src));
#elif defined(ZSTD_ARCH_X86_SSE2)
_mm_storeu_si128((__m128i*)dst, _mm_loadu_si128((const __m128i*)src));
#elif defined(__clang__)
ZSTD_memmove(dst, src, 16);
#else
/* ZSTD_memmove is not inlined properly by gcc */
BYTE copy16_buf[16];
ZSTD_memcpy(copy16_buf, src, 16);
ZSTD_memcpy(dst, copy16_buf, 16);
#endif
}
#define COPY16(d,s) { ZSTD_copy16(d,s); d+=16; s+=16; }
#define WILDCOPY_OVERLENGTH 32
#define WILDCOPY_VECLEN 16
typedef enum {
ZSTD_no_overlap,
ZSTD_overlap_src_before_dst
/* ZSTD_overlap_dst_before_src, */
} ZSTD_overlap_e;
/*! ZSTD_wildcopy() :
* Custom version of ZSTD_memcpy(), can over read/write up to WILDCOPY_OVERLENGTH bytes (if length==0)
* @param ovtype controls the overlap detection
* - ZSTD_no_overlap: The source and destination are guaranteed to be at least WILDCOPY_VECLEN bytes apart.
* - ZSTD_overlap_src_before_dst: The src and dst may overlap, but they MUST be at least 8 bytes apart.
* The src buffer must be before the dst buffer.
*/
MEM_STATIC FORCE_INLINE_ATTR
void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e const ovtype)
{
ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src;
const BYTE* ip = (const BYTE*)src;
BYTE* op = (BYTE*)dst;
BYTE* const oend = op + length;
if (ovtype == ZSTD_overlap_src_before_dst && diff < WILDCOPY_VECLEN) {
/* Handle short offset copies. */
do {
COPY8(op, ip)
} while (op < oend);
} else {
assert(diff >= WILDCOPY_VECLEN || diff <= -WILDCOPY_VECLEN);
/* Separate out the first COPY16() call because the copy length is
* almost certain to be short, so the branches have different
* probabilities. Since it is almost certain to be short, only do
* one COPY16() in the first call. Then, do two calls per loop since
* at that point it is more likely to have a high trip count.
*/
ZSTD_copy16(op, ip);
if (16 >= length) return;
op += 16;
ip += 16;
do {
COPY16(op, ip);
COPY16(op, ip);
}
while (op < oend);
}
}
MEM_STATIC size_t ZSTD_limitCopy(void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
size_t const length = MIN(dstCapacity, srcSize);
if (length > 0) {
ZSTD_memcpy(dst, src, length);
}
return length;
}
/* define "workspace is too large" as this number of times larger than needed */
#define ZSTD_WORKSPACETOOLARGE_FACTOR 3
/* when workspace is continuously too large
* during at least this number of times,
* context's memory usage is considered wasteful,
* because it's sized to handle a worst case scenario which rarely happens.
* In which case, resize it down to free some memory */
#define ZSTD_WORKSPACETOOLARGE_MAXDURATION 128
/* Controls whether the input/output buffer is buffered or stable. */
typedef enum {
ZSTD_bm_buffered = 0, /* Buffer the input/output */
ZSTD_bm_stable = 1 /* ZSTD_inBuffer/ZSTD_outBuffer is stable */
} ZSTD_bufferMode_e;
/*-*******************************************
* Private declarations
*********************************************/
typedef struct seqDef_s {
U32 offBase; /* offBase == Offset + ZSTD_REP_NUM, or repcode 1,2,3 */
U16 litLength;
U16 mlBase; /* mlBase == matchLength - MINMATCH */
} seqDef;
/* Controls whether seqStore has a single "long" litLength or matchLength. See seqStore_t. */
typedef enum {
ZSTD_llt_none = 0, /* no longLengthType */
ZSTD_llt_literalLength = 1, /* represents a long literal */
ZSTD_llt_matchLength = 2 /* represents a long match */
} ZSTD_longLengthType_e;
typedef struct {
seqDef* sequencesStart;
seqDef* sequences; /* ptr to end of sequences */
BYTE* litStart;
BYTE* lit; /* ptr to end of literals */
BYTE* llCode;
BYTE* mlCode;
BYTE* ofCode;
size_t maxNbSeq;
size_t maxNbLit;
/* longLengthPos and longLengthType to allow us to represent either a single litLength or matchLength
* in the seqStore that has a value larger than U16 (if it exists). To do so, we increment
* the existing value of the litLength or matchLength by 0x10000.
*/
ZSTD_longLengthType_e longLengthType;
U32 longLengthPos; /* Index of the sequence to apply long length modification to */
} seqStore_t;
typedef struct {
U32 litLength;
U32 matchLength;
} ZSTD_sequenceLength;
/**
* Returns the ZSTD_sequenceLength for the given sequences. It handles the decoding of long sequences
* indicated by longLengthPos and longLengthType, and adds MINMATCH back to matchLength.
*/
MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore, seqDef const* seq)
{
ZSTD_sequenceLength seqLen;
seqLen.litLength = seq->litLength;
seqLen.matchLength = seq->mlBase + MINMATCH;
if (seqStore->longLengthPos == (U32)(seq - seqStore->sequencesStart)) {
if (seqStore->longLengthType == ZSTD_llt_literalLength) {
seqLen.litLength += 0x10000;
}
if (seqStore->longLengthType == ZSTD_llt_matchLength) {
seqLen.matchLength += 0x10000;
}
}
return seqLen;
}
/**
* Contains the compressed frame size and an upper-bound for the decompressed frame size.
* Note: before using `compressedSize`, check for errors using ZSTD_isError().
* similarly, before using `decompressedBound`, check for errors using:
* `decompressedBound != ZSTD_CONTENTSIZE_ERROR`
*/
typedef struct {
size_t nbBlocks;
size_t compressedSize;
unsigned long long decompressedBound;
} ZSTD_frameSizeInfo; /* decompress & legacy */
const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx); /* compress & dictBuilder */
int ZSTD_seqToCodes(const seqStore_t* seqStorePtr); /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */
/* ZSTD_invalidateRepCodes() :
* ensures next compression will not use repcodes from previous block.
* Note : only works with regular variant;
* do not use with extDict variant ! */
void ZSTD_invalidateRepCodes(ZSTD_CCtx* cctx); /* zstdmt, adaptive_compression (shouldn't get this definition from here) */
typedef struct {
blockType_e blockType;
U32 lastBlock;
U32 origSize;
} blockProperties_t; /* declared here for decompress and fullbench */
/*! ZSTD_getcBlockSize() :
* Provides the size of compressed block from block header `src` */
/* Used by: decompress, fullbench (does not get its definition from here) */
size_t ZSTD_getcBlockSize(const void* src, size_t srcSize,
blockProperties_t* bpPtr);
/*! ZSTD_decodeSeqHeaders() :
* decode sequence header from src */
/* Used by: decompress, fullbench (does not get its definition from here) */
size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr,
const void* src, size_t srcSize);
/**
* @returns true iff the CPU supports dynamic BMI2 dispatch.
*/
MEM_STATIC int ZSTD_cpuSupportsBmi2(void)
{
ZSTD_cpuid_t cpuid = ZSTD_cpuid();
return ZSTD_cpuid_bmi1(cpuid) && ZSTD_cpuid_bmi2(cpuid);
}
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_CCOMMON_H_MODULE */

163
raytracer/nvpro_core/third_party/zstd/lib/common/zstd_trace.h vendored Normal file
View file

@ -0,0 +1,163 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_TRACE_H
#define ZSTD_TRACE_H
#if defined (__cplusplus)
extern "C" {
#endif
#include <stddef.h>
/* weak symbol support
* For now, enable conservatively:
* - Only GNUC
* - Only ELF
* - Only x86-64, i386 and aarch64
* Also, explicitly disable on platforms known not to work so they aren't
* forgotten in the future.
*/
#if !defined(ZSTD_HAVE_WEAK_SYMBOLS) && \
defined(__GNUC__) && defined(__ELF__) && \
(defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || defined(_M_IX86) || defined(__aarch64__)) && \
!defined(__APPLE__) && !defined(_WIN32) && !defined(__MINGW32__) && \
!defined(__CYGWIN__) && !defined(_AIX)
# define ZSTD_HAVE_WEAK_SYMBOLS 1
#else
# define ZSTD_HAVE_WEAK_SYMBOLS 0
#endif
#if ZSTD_HAVE_WEAK_SYMBOLS
# define ZSTD_WEAK_ATTR __attribute__((__weak__))
#else
# define ZSTD_WEAK_ATTR
#endif
/* Only enable tracing when weak symbols are available. */
#ifndef ZSTD_TRACE
# define ZSTD_TRACE ZSTD_HAVE_WEAK_SYMBOLS
#endif
#if ZSTD_TRACE
struct ZSTD_CCtx_s;
struct ZSTD_DCtx_s;
struct ZSTD_CCtx_params_s;
typedef struct {
/**
* ZSTD_VERSION_NUMBER
*
* This is guaranteed to be the first member of ZSTD_trace.
* Otherwise, this struct is not stable between versions. If
* the version number does not match your expectation, you
* should not interpret the rest of the struct.
*/
unsigned version;
/**
* Non-zero if streaming (de)compression is used.
*/
unsigned streaming;
/**
* The dictionary ID.
*/
unsigned dictionaryID;
/**
* Is the dictionary cold?
* Only set on decompression.
*/
unsigned dictionaryIsCold;
/**
* The dictionary size or zero if no dictionary.
*/
size_t dictionarySize;
/**
* The uncompressed size of the data.
*/
size_t uncompressedSize;
/**
* The compressed size of the data.
*/
size_t compressedSize;
/**
* The fully resolved CCtx parameters (NULL on decompression).
*/
struct ZSTD_CCtx_params_s const* params;
/**
* The ZSTD_CCtx pointer (NULL on decompression).
*/
struct ZSTD_CCtx_s const* cctx;
/**
* The ZSTD_DCtx pointer (NULL on compression).
*/
struct ZSTD_DCtx_s const* dctx;
} ZSTD_Trace;
/**
* A tracing context. It must be 0 when tracing is disabled.
* Otherwise, any non-zero value returned by a tracing begin()
* function is presented to any subsequent calls to end().
*
* Any non-zero value is treated as tracing is enabled and not
* interpreted by the library.
*
* Two possible uses are:
* * A timestamp for when the begin() function was called.
* * A unique key identifying the (de)compression, like the
* address of the [dc]ctx pointer if you need to track
* more information than just a timestamp.
*/
typedef unsigned long long ZSTD_TraceCtx;
/**
* Trace the beginning of a compression call.
* @param cctx The dctx pointer for the compression.
* It can be used as a key to map begin() to end().
* @returns Non-zero if tracing is enabled. The return value is
* passed to ZSTD_trace_compress_end().
*/
ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_compress_begin(
struct ZSTD_CCtx_s const* cctx);
/**
* Trace the end of a compression call.
* @param ctx The return value of ZSTD_trace_compress_begin().
* @param trace The zstd tracing info.
*/
ZSTD_WEAK_ATTR void ZSTD_trace_compress_end(
ZSTD_TraceCtx ctx,
ZSTD_Trace const* trace);
/**
* Trace the beginning of a decompression call.
* @param dctx The dctx pointer for the decompression.
* It can be used as a key to map begin() to end().
* @returns Non-zero if tracing is enabled. The return value is
* passed to ZSTD_trace_compress_end().
*/
ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_decompress_begin(
struct ZSTD_DCtx_s const* dctx);
/**
* Trace the end of a decompression call.
* @param ctx The return value of ZSTD_trace_decompress_begin().
* @param trace The zstd tracing info.
*/
ZSTD_WEAK_ATTR void ZSTD_trace_decompress_end(
ZSTD_TraceCtx ctx,
ZSTD_Trace const* trace);
#endif /* ZSTD_TRACE */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_TRACE_H */

134
raytracer/nvpro_core/third_party/zstd/lib/compress/clevels.h vendored Normal file
View file

@ -0,0 +1,134 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_CLEVELS_H
#define ZSTD_CLEVELS_H
#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_compressionParameters */
#include "../zstd.h"
/*-===== Pre-defined compression levels =====-*/
#define ZSTD_MAX_CLEVEL 22
#ifdef __GNUC__
__attribute__((__unused__))
#endif
static const ZSTD_compressionParameters ZSTD_defaultCParameters[4][ZSTD_MAX_CLEVEL+1] = {
{ /* "default" - for any srcSize > 256 KB */
/* W, C, H, S, L, TL, strat */
{ 19, 12, 13, 1, 6, 1, ZSTD_fast }, /* base for negative levels */
{ 19, 13, 14, 1, 7, 0, ZSTD_fast }, /* level 1 */
{ 20, 15, 16, 1, 6, 0, ZSTD_fast }, /* level 2 */
{ 21, 16, 17, 1, 5, 0, ZSTD_dfast }, /* level 3 */
{ 21, 18, 18, 1, 5, 0, ZSTD_dfast }, /* level 4 */
{ 21, 18, 19, 3, 5, 2, ZSTD_greedy }, /* level 5 */
{ 21, 18, 19, 3, 5, 4, ZSTD_lazy }, /* level 6 */
{ 21, 19, 20, 4, 5, 8, ZSTD_lazy }, /* level 7 */
{ 21, 19, 20, 4, 5, 16, ZSTD_lazy2 }, /* level 8 */
{ 22, 20, 21, 4, 5, 16, ZSTD_lazy2 }, /* level 9 */
{ 22, 21, 22, 5, 5, 16, ZSTD_lazy2 }, /* level 10 */
{ 22, 21, 22, 6, 5, 16, ZSTD_lazy2 }, /* level 11 */
{ 22, 22, 23, 6, 5, 32, ZSTD_lazy2 }, /* level 12 */
{ 22, 22, 22, 4, 5, 32, ZSTD_btlazy2 }, /* level 13 */
{ 22, 22, 23, 5, 5, 32, ZSTD_btlazy2 }, /* level 14 */
{ 22, 23, 23, 6, 5, 32, ZSTD_btlazy2 }, /* level 15 */
{ 22, 22, 22, 5, 5, 48, ZSTD_btopt }, /* level 16 */
{ 23, 23, 22, 5, 4, 64, ZSTD_btopt }, /* level 17 */
{ 23, 23, 22, 6, 3, 64, ZSTD_btultra }, /* level 18 */
{ 23, 24, 22, 7, 3,256, ZSTD_btultra2}, /* level 19 */
{ 25, 25, 23, 7, 3,256, ZSTD_btultra2}, /* level 20 */
{ 26, 26, 24, 7, 3,512, ZSTD_btultra2}, /* level 21 */
{ 27, 27, 25, 9, 3,999, ZSTD_btultra2}, /* level 22 */
},
{ /* for srcSize <= 256 KB */
/* W, C, H, S, L, T, strat */
{ 18, 12, 13, 1, 5, 1, ZSTD_fast }, /* base for negative levels */
{ 18, 13, 14, 1, 6, 0, ZSTD_fast }, /* level 1 */
{ 18, 14, 14, 1, 5, 0, ZSTD_dfast }, /* level 2 */
{ 18, 16, 16, 1, 4, 0, ZSTD_dfast }, /* level 3 */
{ 18, 16, 17, 3, 5, 2, ZSTD_greedy }, /* level 4.*/
{ 18, 17, 18, 5, 5, 2, ZSTD_greedy }, /* level 5.*/
{ 18, 18, 19, 3, 5, 4, ZSTD_lazy }, /* level 6.*/
{ 18, 18, 19, 4, 4, 4, ZSTD_lazy }, /* level 7 */
{ 18, 18, 19, 4, 4, 8, ZSTD_lazy2 }, /* level 8 */
{ 18, 18, 19, 5, 4, 8, ZSTD_lazy2 }, /* level 9 */
{ 18, 18, 19, 6, 4, 8, ZSTD_lazy2 }, /* level 10 */
{ 18, 18, 19, 5, 4, 12, ZSTD_btlazy2 }, /* level 11.*/
{ 18, 19, 19, 7, 4, 12, ZSTD_btlazy2 }, /* level 12.*/
{ 18, 18, 19, 4, 4, 16, ZSTD_btopt }, /* level 13 */
{ 18, 18, 19, 4, 3, 32, ZSTD_btopt }, /* level 14.*/
{ 18, 18, 19, 6, 3,128, ZSTD_btopt }, /* level 15.*/
{ 18, 19, 19, 6, 3,128, ZSTD_btultra }, /* level 16.*/
{ 18, 19, 19, 8, 3,256, ZSTD_btultra }, /* level 17.*/
{ 18, 19, 19, 6, 3,128, ZSTD_btultra2}, /* level 18.*/
{ 18, 19, 19, 8, 3,256, ZSTD_btultra2}, /* level 19.*/
{ 18, 19, 19, 10, 3,512, ZSTD_btultra2}, /* level 20.*/
{ 18, 19, 19, 12, 3,512, ZSTD_btultra2}, /* level 21.*/
{ 18, 19, 19, 13, 3,999, ZSTD_btultra2}, /* level 22.*/
},
{ /* for srcSize <= 128 KB */
/* W, C, H, S, L, T, strat */
{ 17, 12, 12, 1, 5, 1, ZSTD_fast }, /* base for negative levels */
{ 17, 12, 13, 1, 6, 0, ZSTD_fast }, /* level 1 */
{ 17, 13, 15, 1, 5, 0, ZSTD_fast }, /* level 2 */
{ 17, 15, 16, 2, 5, 0, ZSTD_dfast }, /* level 3 */
{ 17, 17, 17, 2, 4, 0, ZSTD_dfast }, /* level 4 */
{ 17, 16, 17, 3, 4, 2, ZSTD_greedy }, /* level 5 */
{ 17, 16, 17, 3, 4, 4, ZSTD_lazy }, /* level 6 */
{ 17, 16, 17, 3, 4, 8, ZSTD_lazy2 }, /* level 7 */
{ 17, 16, 17, 4, 4, 8, ZSTD_lazy2 }, /* level 8 */
{ 17, 16, 17, 5, 4, 8, ZSTD_lazy2 }, /* level 9 */
{ 17, 16, 17, 6, 4, 8, ZSTD_lazy2 }, /* level 10 */
{ 17, 17, 17, 5, 4, 8, ZSTD_btlazy2 }, /* level 11 */
{ 17, 18, 17, 7, 4, 12, ZSTD_btlazy2 }, /* level 12 */
{ 17, 18, 17, 3, 4, 12, ZSTD_btopt }, /* level 13.*/
{ 17, 18, 17, 4, 3, 32, ZSTD_btopt }, /* level 14.*/
{ 17, 18, 17, 6, 3,256, ZSTD_btopt }, /* level 15.*/
{ 17, 18, 17, 6, 3,128, ZSTD_btultra }, /* level 16.*/
{ 17, 18, 17, 8, 3,256, ZSTD_btultra }, /* level 17.*/
{ 17, 18, 17, 10, 3,512, ZSTD_btultra }, /* level 18.*/
{ 17, 18, 17, 5, 3,256, ZSTD_btultra2}, /* level 19.*/
{ 17, 18, 17, 7, 3,512, ZSTD_btultra2}, /* level 20.*/
{ 17, 18, 17, 9, 3,512, ZSTD_btultra2}, /* level 21.*/
{ 17, 18, 17, 11, 3,999, ZSTD_btultra2}, /* level 22.*/
},
{ /* for srcSize <= 16 KB */
/* W, C, H, S, L, T, strat */
{ 14, 12, 13, 1, 5, 1, ZSTD_fast }, /* base for negative levels */
{ 14, 14, 15, 1, 5, 0, ZSTD_fast }, /* level 1 */
{ 14, 14, 15, 1, 4, 0, ZSTD_fast }, /* level 2 */
{ 14, 14, 15, 2, 4, 0, ZSTD_dfast }, /* level 3 */
{ 14, 14, 14, 4, 4, 2, ZSTD_greedy }, /* level 4 */
{ 14, 14, 14, 3, 4, 4, ZSTD_lazy }, /* level 5.*/
{ 14, 14, 14, 4, 4, 8, ZSTD_lazy2 }, /* level 6 */
{ 14, 14, 14, 6, 4, 8, ZSTD_lazy2 }, /* level 7 */
{ 14, 14, 14, 8, 4, 8, ZSTD_lazy2 }, /* level 8.*/
{ 14, 15, 14, 5, 4, 8, ZSTD_btlazy2 }, /* level 9.*/
{ 14, 15, 14, 9, 4, 8, ZSTD_btlazy2 }, /* level 10.*/
{ 14, 15, 14, 3, 4, 12, ZSTD_btopt }, /* level 11.*/
{ 14, 15, 14, 4, 3, 24, ZSTD_btopt }, /* level 12.*/
{ 14, 15, 14, 5, 3, 32, ZSTD_btultra }, /* level 13.*/
{ 14, 15, 15, 6, 3, 64, ZSTD_btultra }, /* level 14.*/
{ 14, 15, 15, 7, 3,256, ZSTD_btultra }, /* level 15.*/
{ 14, 15, 15, 5, 3, 48, ZSTD_btultra2}, /* level 16.*/
{ 14, 15, 15, 6, 3,128, ZSTD_btultra2}, /* level 17.*/
{ 14, 15, 15, 7, 3,256, ZSTD_btultra2}, /* level 18.*/
{ 14, 15, 15, 8, 3,256, ZSTD_btultra2}, /* level 19.*/
{ 14, 15, 15, 8, 3,512, ZSTD_btultra2}, /* level 20.*/
{ 14, 15, 15, 9, 3,512, ZSTD_btultra2}, /* level 21.*/
{ 14, 15, 15, 10, 3,999, ZSTD_btultra2}, /* level 22.*/
},
};
#endif /* ZSTD_CLEVELS_H */

624
raytracer/nvpro_core/third_party/zstd/lib/compress/fse_compress.c vendored Normal file
View file

@ -0,0 +1,624 @@
/* ******************************************************************
* FSE : Finite State Entropy encoder
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* **************************************************************
* Includes
****************************************************************/
#include "../common/compiler.h"
#include "../common/mem.h" /* U32, U16, etc. */
#include "../common/debug.h" /* assert, DEBUGLOG */
#include "hist.h" /* HIST_count_wksp */
#include "../common/bitstream.h"
#define FSE_STATIC_LINKING_ONLY
#include "../common/fse.h"
#include "../common/error_private.h"
#define ZSTD_DEPS_NEED_MALLOC
#define ZSTD_DEPS_NEED_MATH64
#include "../common/zstd_deps.h" /* ZSTD_malloc, ZSTD_free, ZSTD_memcpy, ZSTD_memset */
#include "../common/bits.h" /* ZSTD_highbit32 */
/* **************************************************************
* Error Management
****************************************************************/
#define FSE_isError ERR_isError
/* **************************************************************
* Templates
****************************************************************/
/*
designed to be included
for type-specific functions (template emulation in C)
Objective is to write these functions only once, for improved maintenance
*/
/* safety checks */
#ifndef FSE_FUNCTION_EXTENSION
# error "FSE_FUNCTION_EXTENSION must be defined"
#endif
#ifndef FSE_FUNCTION_TYPE
# error "FSE_FUNCTION_TYPE must be defined"
#endif
/* Function names */
#define FSE_CAT(X,Y) X##Y
#define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y)
#define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y)
/* Function templates */
/* FSE_buildCTable_wksp() :
* Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`).
* wkspSize should be sized to handle worst case situation, which is `1<<max_tableLog * sizeof(FSE_FUNCTION_TYPE)`
* workSpace must also be properly aligned with FSE_FUNCTION_TYPE requirements
*/
size_t FSE_buildCTable_wksp(FSE_CTable* ct,
const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog,
void* workSpace, size_t wkspSize)
{
U32 const tableSize = 1 << tableLog;
U32 const tableMask = tableSize - 1;
void* const ptr = ct;
U16* const tableU16 = ( (U16*) ptr) + 2;
void* const FSCT = ((U32*)ptr) + 1 /* header */ + (tableLog ? tableSize>>1 : 1) ;
FSE_symbolCompressionTransform* const symbolTT = (FSE_symbolCompressionTransform*) (FSCT);
U32 const step = FSE_TABLESTEP(tableSize);
U32 const maxSV1 = maxSymbolValue+1;
U16* cumul = (U16*)workSpace; /* size = maxSV1 */
FSE_FUNCTION_TYPE* const tableSymbol = (FSE_FUNCTION_TYPE*)(cumul + (maxSV1+1)); /* size = tableSize */
U32 highThreshold = tableSize-1;
assert(((size_t)workSpace & 1) == 0); /* Must be 2 bytes-aligned */
if (FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) > wkspSize) return ERROR(tableLog_tooLarge);
/* CTable header */
tableU16[-2] = (U16) tableLog;
tableU16[-1] = (U16) maxSymbolValue;
assert(tableLog < 16); /* required for threshold strategy to work */
/* For explanations on how to distribute symbol values over the table :
* https://fastcompression.blogspot.fr/2014/02/fse-distributing-symbol-values.html */
#ifdef __clang_analyzer__
ZSTD_memset(tableSymbol, 0, sizeof(*tableSymbol) * tableSize); /* useless initialization, just to keep scan-build happy */
#endif
/* symbol start positions */
{ U32 u;
cumul[0] = 0;
for (u=1; u <= maxSV1; u++) {
if (normalizedCounter[u-1]==-1) { /* Low proba symbol */
cumul[u] = cumul[u-1] + 1;
tableSymbol[highThreshold--] = (FSE_FUNCTION_TYPE)(u-1);
} else {
assert(normalizedCounter[u-1] >= 0);
cumul[u] = cumul[u-1] + (U16)normalizedCounter[u-1];
assert(cumul[u] >= cumul[u-1]); /* no overflow */
} }
cumul[maxSV1] = (U16)(tableSize+1);
}
/* Spread symbols */
if (highThreshold == tableSize - 1) {
/* Case for no low prob count symbols. Lay down 8 bytes at a time
* to reduce branch misses since we are operating on a small block
*/
BYTE* const spread = tableSymbol + tableSize; /* size = tableSize + 8 (may write beyond tableSize) */
{ U64 const add = 0x0101010101010101ull;
size_t pos = 0;
U64 sv = 0;
U32 s;
for (s=0; s<maxSV1; ++s, sv += add) {
int i;
int const n = normalizedCounter[s];
MEM_write64(spread + pos, sv);
for (i = 8; i < n; i += 8) {
MEM_write64(spread + pos + i, sv);
}
assert(n>=0);
pos += (size_t)n;
}
}
/* Spread symbols across the table. Lack of lowprob symbols means that
* we don't need variable sized inner loop, so we can unroll the loop and
* reduce branch misses.
*/
{ size_t position = 0;
size_t s;
size_t const unroll = 2; /* Experimentally determined optimal unroll */
assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */
for (s = 0; s < (size_t)tableSize; s += unroll) {
size_t u;
for (u = 0; u < unroll; ++u) {
size_t const uPosition = (position + (u * step)) & tableMask;
tableSymbol[uPosition] = spread[s + u];
}
position = (position + (unroll * step)) & tableMask;
}
assert(position == 0); /* Must have initialized all positions */
}
} else {
U32 position = 0;
U32 symbol;
for (symbol=0; symbol<maxSV1; symbol++) {
int nbOccurrences;
int const freq = normalizedCounter[symbol];
for (nbOccurrences=0; nbOccurrences<freq; nbOccurrences++) {
tableSymbol[position] = (FSE_FUNCTION_TYPE)symbol;
position = (position + step) & tableMask;
while (position > highThreshold)
position = (position + step) & tableMask; /* Low proba area */
} }
assert(position==0); /* Must have initialized all positions */
}
/* Build table */
{ U32 u; for (u=0; u<tableSize; u++) {
FSE_FUNCTION_TYPE s = tableSymbol[u]; /* note : static analyzer may not understand tableSymbol is properly initialized */
tableU16[cumul[s]++] = (U16) (tableSize+u); /* TableU16 : sorted by symbol order; gives next state value */
} }
/* Build Symbol Transformation Table */
{ unsigned total = 0;
unsigned s;
for (s=0; s<=maxSymbolValue; s++) {
switch (normalizedCounter[s])
{
case 0:
/* filling nonetheless, for compatibility with FSE_getMaxNbBits() */
symbolTT[s].deltaNbBits = ((tableLog+1) << 16) - (1<<tableLog);
break;
case -1:
case 1:
symbolTT[s].deltaNbBits = (tableLog << 16) - (1<<tableLog);
assert(total <= INT_MAX);
symbolTT[s].deltaFindState = (int)(total - 1);
total ++;
break;
default :
assert(normalizedCounter[s] > 1);
{ U32 const maxBitsOut = tableLog - ZSTD_highbit32 ((U32)normalizedCounter[s]-1);
U32 const minStatePlus = (U32)normalizedCounter[s] << maxBitsOut;
symbolTT[s].deltaNbBits = (maxBitsOut << 16) - minStatePlus;
symbolTT[s].deltaFindState = (int)(total - (unsigned)normalizedCounter[s]);
total += (unsigned)normalizedCounter[s];
} } } }
#if 0 /* debug : symbol costs */
DEBUGLOG(5, "\n --- table statistics : ");
{ U32 symbol;
for (symbol=0; symbol<=maxSymbolValue; symbol++) {
DEBUGLOG(5, "%3u: w=%3i, maxBits=%u, fracBits=%.2f",
symbol, normalizedCounter[symbol],
FSE_getMaxNbBits(symbolTT, symbol),
(double)FSE_bitCost(symbolTT, tableLog, symbol, 8) / 256);
} }
#endif
return 0;
}
#ifndef FSE_COMMONDEFS_ONLY
/*-**************************************************************
* FSE NCount encoding
****************************************************************/
size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog)
{
size_t const maxHeaderSize = (((maxSymbolValue+1) * tableLog
+ 4 /* bitCount initialized at 4 */
+ 2 /* first two symbols may use one additional bit each */) / 8)
+ 1 /* round up to whole nb bytes */
+ 2 /* additional two bytes for bitstream flush */;
return maxSymbolValue ? maxHeaderSize : FSE_NCOUNTBOUND; /* maxSymbolValue==0 ? use default */
}
static size_t
FSE_writeNCount_generic (void* header, size_t headerBufferSize,
const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog,
unsigned writeIsSafe)
{
BYTE* const ostart = (BYTE*) header;
BYTE* out = ostart;
BYTE* const oend = ostart + headerBufferSize;
int nbBits;
const int tableSize = 1 << tableLog;
int remaining;
int threshold;
U32 bitStream = 0;
int bitCount = 0;
unsigned symbol = 0;
unsigned const alphabetSize = maxSymbolValue + 1;
int previousIs0 = 0;
/* Table Size */
bitStream += (tableLog-FSE_MIN_TABLELOG) << bitCount;
bitCount += 4;
/* Init */
remaining = tableSize+1; /* +1 for extra accuracy */
threshold = tableSize;
nbBits = tableLog+1;
while ((symbol < alphabetSize) && (remaining>1)) { /* stops at 1 */
if (previousIs0) {
unsigned start = symbol;
while ((symbol < alphabetSize) && !normalizedCounter[symbol]) symbol++;
if (symbol == alphabetSize) break; /* incorrect distribution */
while (symbol >= start+24) {
start+=24;
bitStream += 0xFFFFU << bitCount;
if ((!writeIsSafe) && (out > oend-2))
return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE) bitStream;
out[1] = (BYTE)(bitStream>>8);
out+=2;
bitStream>>=16;
}
while (symbol >= start+3) {
start+=3;
bitStream += 3 << bitCount;
bitCount += 2;
}
bitStream += (symbol-start) << bitCount;
bitCount += 2;
if (bitCount>16) {
if ((!writeIsSafe) && (out > oend - 2))
return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE)bitStream;
out[1] = (BYTE)(bitStream>>8);
out += 2;
bitStream >>= 16;
bitCount -= 16;
} }
{ int count = normalizedCounter[symbol++];
int const max = (2*threshold-1) - remaining;
remaining -= count < 0 ? -count : count;
count++; /* +1 for extra accuracy */
if (count>=threshold)
count += max; /* [0..max[ [max..threshold[ (...) [threshold+max 2*threshold[ */
bitStream += count << bitCount;
bitCount += nbBits;
bitCount -= (count<max);
previousIs0 = (count==1);
if (remaining<1) return ERROR(GENERIC);
while (remaining<threshold) { nbBits--; threshold>>=1; }
}
if (bitCount>16) {
if ((!writeIsSafe) && (out > oend - 2))
return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE)bitStream;
out[1] = (BYTE)(bitStream>>8);
out += 2;
bitStream >>= 16;
bitCount -= 16;
} }
if (remaining != 1)
return ERROR(GENERIC); /* incorrect normalized distribution */
assert(symbol <= alphabetSize);
/* flush remaining bitStream */
if ((!writeIsSafe) && (out > oend - 2))
return ERROR(dstSize_tooSmall); /* Buffer overflow */
out[0] = (BYTE)bitStream;
out[1] = (BYTE)(bitStream>>8);
out+= (bitCount+7) /8;
return (out-ostart);
}
size_t FSE_writeNCount (void* buffer, size_t bufferSize,
const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog)
{
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); /* Unsupported */
if (tableLog < FSE_MIN_TABLELOG) return ERROR(GENERIC); /* Unsupported */
if (bufferSize < FSE_NCountWriteBound(maxSymbolValue, tableLog))
return FSE_writeNCount_generic(buffer, bufferSize, normalizedCounter, maxSymbolValue, tableLog, 0);
return FSE_writeNCount_generic(buffer, bufferSize, normalizedCounter, maxSymbolValue, tableLog, 1 /* write in buffer is safe */);
}
/*-**************************************************************
* FSE Compression Code
****************************************************************/
/* provides the minimum logSize to safely represent a distribution */
static unsigned FSE_minTableLog(size_t srcSize, unsigned maxSymbolValue)
{
U32 minBitsSrc = ZSTD_highbit32((U32)(srcSize)) + 1;
U32 minBitsSymbols = ZSTD_highbit32(maxSymbolValue) + 2;
U32 minBits = minBitsSrc < minBitsSymbols ? minBitsSrc : minBitsSymbols;
assert(srcSize > 1); /* Not supported, RLE should be used instead */
return minBits;
}
unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus)
{
U32 maxBitsSrc = ZSTD_highbit32((U32)(srcSize - 1)) - minus;
U32 tableLog = maxTableLog;
U32 minBits = FSE_minTableLog(srcSize, maxSymbolValue);
assert(srcSize > 1); /* Not supported, RLE should be used instead */
if (tableLog==0) tableLog = FSE_DEFAULT_TABLELOG;
if (maxBitsSrc < tableLog) tableLog = maxBitsSrc; /* Accuracy can be reduced */
if (minBits > tableLog) tableLog = minBits; /* Need a minimum to safely represent all symbol values */
if (tableLog < FSE_MIN_TABLELOG) tableLog = FSE_MIN_TABLELOG;
if (tableLog > FSE_MAX_TABLELOG) tableLog = FSE_MAX_TABLELOG;
return tableLog;
}
unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue)
{
return FSE_optimalTableLog_internal(maxTableLog, srcSize, maxSymbolValue, 2);
}
/* Secondary normalization method.
To be used when primary method fails. */
static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, size_t total, U32 maxSymbolValue, short lowProbCount)
{
short const NOT_YET_ASSIGNED = -2;
U32 s;
U32 distributed = 0;
U32 ToDistribute;
/* Init */
U32 const lowThreshold = (U32)(total >> tableLog);
U32 lowOne = (U32)((total * 3) >> (tableLog + 1));
for (s=0; s<=maxSymbolValue; s++) {
if (count[s] == 0) {
norm[s]=0;
continue;
}
if (count[s] <= lowThreshold) {
norm[s] = lowProbCount;
distributed++;
total -= count[s];
continue;
}
if (count[s] <= lowOne) {
norm[s] = 1;
distributed++;
total -= count[s];
continue;
}
norm[s]=NOT_YET_ASSIGNED;
}
ToDistribute = (1 << tableLog) - distributed;
if (ToDistribute == 0)
return 0;
if ((total / ToDistribute) > lowOne) {
/* risk of rounding to zero */
lowOne = (U32)((total * 3) / (ToDistribute * 2));
for (s=0; s<=maxSymbolValue; s++) {
if ((norm[s] == NOT_YET_ASSIGNED) && (count[s] <= lowOne)) {
norm[s] = 1;
distributed++;
total -= count[s];
continue;
} }
ToDistribute = (1 << tableLog) - distributed;
}
if (distributed == maxSymbolValue+1) {
/* all values are pretty poor;
probably incompressible data (should have already been detected);
find max, then give all remaining points to max */
U32 maxV = 0, maxC = 0;
for (s=0; s<=maxSymbolValue; s++)
if (count[s] > maxC) { maxV=s; maxC=count[s]; }
norm[maxV] += (short)ToDistribute;
return 0;
}
if (total == 0) {
/* all of the symbols were low enough for the lowOne or lowThreshold */
for (s=0; ToDistribute > 0; s = (s+1)%(maxSymbolValue+1))
if (norm[s] > 0) { ToDistribute--; norm[s]++; }
return 0;
}
{ U64 const vStepLog = 62 - tableLog;
U64 const mid = (1ULL << (vStepLog-1)) - 1;
U64 const rStep = ZSTD_div64((((U64)1<<vStepLog) * ToDistribute) + mid, (U32)total); /* scale on remaining */
U64 tmpTotal = mid;
for (s=0; s<=maxSymbolValue; s++) {
if (norm[s]==NOT_YET_ASSIGNED) {
U64 const end = tmpTotal + (count[s] * rStep);
U32 const sStart = (U32)(tmpTotal >> vStepLog);
U32 const sEnd = (U32)(end >> vStepLog);
U32 const weight = sEnd - sStart;
if (weight < 1)
return ERROR(GENERIC);
norm[s] = (short)weight;
tmpTotal = end;
} } }
return 0;
}
size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog,
const unsigned* count, size_t total,
unsigned maxSymbolValue, unsigned useLowProbCount)
{
/* Sanity checks */
if (tableLog==0) tableLog = FSE_DEFAULT_TABLELOG;
if (tableLog < FSE_MIN_TABLELOG) return ERROR(GENERIC); /* Unsupported size */
if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); /* Unsupported size */
if (tableLog < FSE_minTableLog(total, maxSymbolValue)) return ERROR(GENERIC); /* Too small tableLog, compression potentially impossible */
{ static U32 const rtbTable[] = { 0, 473195, 504333, 520860, 550000, 700000, 750000, 830000 };
short const lowProbCount = useLowProbCount ? -1 : 1;
U64 const scale = 62 - tableLog;
U64 const step = ZSTD_div64((U64)1<<62, (U32)total); /* <== here, one division ! */
U64 const vStep = 1ULL<<(scale-20);
int stillToDistribute = 1<<tableLog;
unsigned s;
unsigned largest=0;
short largestP=0;
U32 lowThreshold = (U32)(total >> tableLog);
for (s=0; s<=maxSymbolValue; s++) {
if (count[s] == total) return 0; /* rle special case */
if (count[s] == 0) { normalizedCounter[s]=0; continue; }
if (count[s] <= lowThreshold) {
normalizedCounter[s] = lowProbCount;
stillToDistribute--;
} else {
short proba = (short)((count[s]*step) >> scale);
if (proba<8) {
U64 restToBeat = vStep * rtbTable[proba];
proba += (count[s]*step) - ((U64)proba<<scale) > restToBeat;
}
if (proba > largestP) { largestP=proba; largest=s; }
normalizedCounter[s] = proba;
stillToDistribute -= proba;
} }
if (-stillToDistribute >= (normalizedCounter[largest] >> 1)) {
/* corner case, need another normalization method */
size_t const errorCode = FSE_normalizeM2(normalizedCounter, tableLog, count, total, maxSymbolValue, lowProbCount);
if (FSE_isError(errorCode)) return errorCode;
}
else normalizedCounter[largest] += (short)stillToDistribute;
}
#if 0
{ /* Print Table (debug) */
U32 s;
U32 nTotal = 0;
for (s=0; s<=maxSymbolValue; s++)
RAWLOG(2, "%3i: %4i \n", s, normalizedCounter[s]);
for (s=0; s<=maxSymbolValue; s++)
nTotal += abs(normalizedCounter[s]);
if (nTotal != (1U<<tableLog))
RAWLOG(2, "Warning !!! Total == %u != %u !!!", nTotal, 1U<<tableLog);
getchar();
}
#endif
return tableLog;
}
/* fake FSE_CTable, for rle input (always same symbol) */
size_t FSE_buildCTable_rle (FSE_CTable* ct, BYTE symbolValue)
{
void* ptr = ct;
U16* tableU16 = ( (U16*) ptr) + 2;
void* FSCTptr = (U32*)ptr + 2;
FSE_symbolCompressionTransform* symbolTT = (FSE_symbolCompressionTransform*) FSCTptr;
/* header */
tableU16[-2] = (U16) 0;
tableU16[-1] = (U16) symbolValue;
/* Build table */
tableU16[0] = 0;
tableU16[1] = 0; /* just in case */
/* Build Symbol Transformation Table */
symbolTT[symbolValue].deltaNbBits = 0;
symbolTT[symbolValue].deltaFindState = 0;
return 0;
}
static size_t FSE_compress_usingCTable_generic (void* dst, size_t dstSize,
const void* src, size_t srcSize,
const FSE_CTable* ct, const unsigned fast)
{
const BYTE* const istart = (const BYTE*) src;
const BYTE* const iend = istart + srcSize;
const BYTE* ip=iend;
BIT_CStream_t bitC;
FSE_CState_t CState1, CState2;
/* init */
if (srcSize <= 2) return 0;
{ size_t const initError = BIT_initCStream(&bitC, dst, dstSize);
if (FSE_isError(initError)) return 0; /* not enough space available to write a bitstream */ }
#define FSE_FLUSHBITS(s) (fast ? BIT_flushBitsFast(s) : BIT_flushBits(s))
if (srcSize & 1) {
FSE_initCState2(&CState1, ct, *--ip);
FSE_initCState2(&CState2, ct, *--ip);
FSE_encodeSymbol(&bitC, &CState1, *--ip);
FSE_FLUSHBITS(&bitC);
} else {
FSE_initCState2(&CState2, ct, *--ip);
FSE_initCState2(&CState1, ct, *--ip);
}
/* join to mod 4 */
srcSize -= 2;
if ((sizeof(bitC.bitContainer)*8 > FSE_MAX_TABLELOG*4+7 ) && (srcSize & 2)) { /* test bit 2 */
FSE_encodeSymbol(&bitC, &CState2, *--ip);
FSE_encodeSymbol(&bitC, &CState1, *--ip);
FSE_FLUSHBITS(&bitC);
}
/* 2 or 4 encoding per loop */
while ( ip>istart ) {
FSE_encodeSymbol(&bitC, &CState2, *--ip);
if (sizeof(bitC.bitContainer)*8 < FSE_MAX_TABLELOG*2+7 ) /* this test must be static */
FSE_FLUSHBITS(&bitC);
FSE_encodeSymbol(&bitC, &CState1, *--ip);
if (sizeof(bitC.bitContainer)*8 > FSE_MAX_TABLELOG*4+7 ) { /* this test must be static */
FSE_encodeSymbol(&bitC, &CState2, *--ip);
FSE_encodeSymbol(&bitC, &CState1, *--ip);
}
FSE_FLUSHBITS(&bitC);
}
FSE_flushCState(&bitC, &CState2);
FSE_flushCState(&bitC, &CState1);
return BIT_closeCStream(&bitC);
}
size_t FSE_compress_usingCTable (void* dst, size_t dstSize,
const void* src, size_t srcSize,
const FSE_CTable* ct)
{
unsigned const fast = (dstSize >= FSE_BLOCKBOUND(srcSize));
if (fast)
return FSE_compress_usingCTable_generic(dst, dstSize, src, srcSize, ct, 1);
else
return FSE_compress_usingCTable_generic(dst, dstSize, src, srcSize, ct, 0);
}
size_t FSE_compressBound(size_t size) { return FSE_COMPRESSBOUND(size); }
#endif /* FSE_COMMONDEFS_ONLY */

181
raytracer/nvpro_core/third_party/zstd/lib/compress/hist.c vendored Normal file
View file

@ -0,0 +1,181 @@
/* ******************************************************************
* hist : Histogram functions
* part of Finite State Entropy project
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* --- dependencies --- */
#include "../common/mem.h" /* U32, BYTE, etc. */
#include "../common/debug.h" /* assert, DEBUGLOG */
#include "../common/error_private.h" /* ERROR */
#include "hist.h"
/* --- Error management --- */
unsigned HIST_isError(size_t code) { return ERR_isError(code); }
/*-**************************************************************
* Histogram functions
****************************************************************/
unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize)
{
const BYTE* ip = (const BYTE*)src;
const BYTE* const end = ip + srcSize;
unsigned maxSymbolValue = *maxSymbolValuePtr;
unsigned largestCount=0;
ZSTD_memset(count, 0, (maxSymbolValue+1) * sizeof(*count));
if (srcSize==0) { *maxSymbolValuePtr = 0; return 0; }
while (ip<end) {
assert(*ip <= maxSymbolValue);
count[*ip++]++;
}
while (!count[maxSymbolValue]) maxSymbolValue--;
*maxSymbolValuePtr = maxSymbolValue;
{ U32 s;
for (s=0; s<=maxSymbolValue; s++)
if (count[s] > largestCount) largestCount = count[s];
}
return largestCount;
}
typedef enum { trustInput, checkMaxSymbolValue } HIST_checkInput_e;
/* HIST_count_parallel_wksp() :
* store histogram into 4 intermediate tables, recombined at the end.
* this design makes better use of OoO cpus,
* and is noticeably faster when some values are heavily repeated.
* But it needs some additional workspace for intermediate tables.
* `workSpace` must be a U32 table of size >= HIST_WKSP_SIZE_U32.
* @return : largest histogram frequency,
* or an error code (notably when histogram's alphabet is larger than *maxSymbolValuePtr) */
static size_t HIST_count_parallel_wksp(
unsigned* count, unsigned* maxSymbolValuePtr,
const void* source, size_t sourceSize,
HIST_checkInput_e check,
U32* const workSpace)
{
const BYTE* ip = (const BYTE*)source;
const BYTE* const iend = ip+sourceSize;
size_t const countSize = (*maxSymbolValuePtr + 1) * sizeof(*count);
unsigned max=0;
U32* const Counting1 = workSpace;
U32* const Counting2 = Counting1 + 256;
U32* const Counting3 = Counting2 + 256;
U32* const Counting4 = Counting3 + 256;
/* safety checks */
assert(*maxSymbolValuePtr <= 255);
if (!sourceSize) {
ZSTD_memset(count, 0, countSize);
*maxSymbolValuePtr = 0;
return 0;
}
ZSTD_memset(workSpace, 0, 4*256*sizeof(unsigned));
/* by stripes of 16 bytes */
{ U32 cached = MEM_read32(ip); ip += 4;
while (ip < iend-15) {
U32 c = cached; cached = MEM_read32(ip); ip += 4;
Counting1[(BYTE) c ]++;
Counting2[(BYTE)(c>>8) ]++;
Counting3[(BYTE)(c>>16)]++;
Counting4[ c>>24 ]++;
c = cached; cached = MEM_read32(ip); ip += 4;
Counting1[(BYTE) c ]++;
Counting2[(BYTE)(c>>8) ]++;
Counting3[(BYTE)(c>>16)]++;
Counting4[ c>>24 ]++;
c = cached; cached = MEM_read32(ip); ip += 4;
Counting1[(BYTE) c ]++;
Counting2[(BYTE)(c>>8) ]++;
Counting3[(BYTE)(c>>16)]++;
Counting4[ c>>24 ]++;
c = cached; cached = MEM_read32(ip); ip += 4;
Counting1[(BYTE) c ]++;
Counting2[(BYTE)(c>>8) ]++;
Counting3[(BYTE)(c>>16)]++;
Counting4[ c>>24 ]++;
}
ip-=4;
}
/* finish last symbols */
while (ip<iend) Counting1[*ip++]++;
{ U32 s;
for (s=0; s<256; s++) {
Counting1[s] += Counting2[s] + Counting3[s] + Counting4[s];
if (Counting1[s] > max) max = Counting1[s];
} }
{ unsigned maxSymbolValue = 255;
while (!Counting1[maxSymbolValue]) maxSymbolValue--;
if (check && maxSymbolValue > *maxSymbolValuePtr) return ERROR(maxSymbolValue_tooSmall);
*maxSymbolValuePtr = maxSymbolValue;
ZSTD_memmove(count, Counting1, countSize); /* in case count & Counting1 are overlapping */
}
return (size_t)max;
}
/* HIST_countFast_wksp() :
* Same as HIST_countFast(), but using an externally provided scratch buffer.
* `workSpace` is a writable buffer which must be 4-bytes aligned,
* `workSpaceSize` must be >= HIST_WKSP_SIZE
*/
size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
const void* source, size_t sourceSize,
void* workSpace, size_t workSpaceSize)
{
if (sourceSize < 1500) /* heuristic threshold */
return HIST_count_simple(count, maxSymbolValuePtr, source, sourceSize);
if ((size_t)workSpace & 3) return ERROR(GENERIC); /* must be aligned on 4-bytes boundaries */
if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall);
return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, trustInput, (U32*)workSpace);
}
/* HIST_count_wksp() :
* Same as HIST_count(), but using an externally provided scratch buffer.
* `workSpace` size must be table of >= HIST_WKSP_SIZE_U32 unsigned */
size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
const void* source, size_t sourceSize,
void* workSpace, size_t workSpaceSize)
{
if ((size_t)workSpace & 3) return ERROR(GENERIC); /* must be aligned on 4-bytes boundaries */
if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall);
if (*maxSymbolValuePtr < 255)
return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, checkMaxSymbolValue, (U32*)workSpace);
*maxSymbolValuePtr = 255;
return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, workSpace, workSpaceSize);
}
#ifndef ZSTD_NO_UNUSED_FUNCTIONS
/* fast variant (unsafe : won't check if src contains values beyond count[] limit) */
size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr,
const void* source, size_t sourceSize)
{
unsigned tmpCounters[HIST_WKSP_SIZE_U32];
return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, tmpCounters, sizeof(tmpCounters));
}
size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize)
{
unsigned tmpCounters[HIST_WKSP_SIZE_U32];
return HIST_count_wksp(count, maxSymbolValuePtr, src, srcSize, tmpCounters, sizeof(tmpCounters));
}
#endif

75
raytracer/nvpro_core/third_party/zstd/lib/compress/hist.h vendored Normal file
View file

@ -0,0 +1,75 @@
/* ******************************************************************
* hist : Histogram functions
* part of Finite State Entropy project
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* You can contact the author at :
* - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
* - Public forum : https://groups.google.com/forum/#!forum/lz4c
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
****************************************************************** */
/* --- dependencies --- */
#include "../common/zstd_deps.h" /* size_t */
/* --- simple histogram functions --- */
/*! HIST_count():
* Provides the precise count of each byte within a table 'count'.
* 'count' is a table of unsigned int, of minimum size (*maxSymbolValuePtr+1).
* Updates *maxSymbolValuePtr with actual largest symbol value detected.
* @return : count of the most frequent symbol (which isn't identified).
* or an error code, which can be tested using HIST_isError().
* note : if return == srcSize, there is only one symbol.
*/
size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize);
unsigned HIST_isError(size_t code); /**< tells if a return value is an error code */
/* --- advanced histogram functions --- */
#define HIST_WKSP_SIZE_U32 1024
#define HIST_WKSP_SIZE (HIST_WKSP_SIZE_U32 * sizeof(unsigned))
/** HIST_count_wksp() :
* Same as HIST_count(), but using an externally provided scratch buffer.
* Benefit is this function will use very little stack space.
* `workSpace` is a writable buffer which must be 4-bytes aligned,
* `workSpaceSize` must be >= HIST_WKSP_SIZE
*/
size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize,
void* workSpace, size_t workSpaceSize);
/** HIST_countFast() :
* same as HIST_count(), but blindly trusts that all byte values within src are <= *maxSymbolValuePtr.
* This function is unsafe, and will segfault if any value within `src` is `> *maxSymbolValuePtr`
*/
size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize);
/** HIST_countFast_wksp() :
* Same as HIST_countFast(), but using an externally provided scratch buffer.
* `workSpace` is a writable buffer which must be 4-bytes aligned,
* `workSpaceSize` must be >= HIST_WKSP_SIZE
*/
size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize,
void* workSpace, size_t workSpaceSize);
/*! HIST_count_simple() :
* Same as HIST_countFast(), this function is unsafe,
* and will segfault if any value within `src` is `> *maxSymbolValuePtr`.
* It is also a bit slower for large inputs.
* However, it does not need any additional memory (not even on stack).
* @return : count of the most frequent symbol.
* Note this function doesn't produce any error (i.e. it must succeed).
*/
unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
const void* src, size_t srcSize);

1435
raytracer/nvpro_core/third_party/zstd/lib/compress/huf_compress.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

7032
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_compress.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

1532
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_compress_internal.h vendored Normal file
View file

File diff suppressed because it is too large Load diff

235
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_compress_literals.c vendored Normal file
View file

@ -0,0 +1,235 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*-*************************************
* Dependencies
***************************************/
#include "zstd_compress_literals.h"
/* **************************************************************
* Debug Traces
****************************************************************/
#if DEBUGLEVEL >= 2
static size_t showHexa(const void* src, size_t srcSize)
{
const BYTE* const ip = (const BYTE*)src;
size_t u;
for (u=0; u<srcSize; u++) {
RAWLOG(5, " %02X", ip[u]); (void)ip;
}
RAWLOG(5, " \n");
return srcSize;
}
#endif
/* **************************************************************
* Literals compression - special cases
****************************************************************/
size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
BYTE* const ostart = (BYTE*)dst;
U32 const flSize = 1 + (srcSize>31) + (srcSize>4095);
DEBUGLOG(5, "ZSTD_noCompressLiterals: srcSize=%zu, dstCapacity=%zu", srcSize, dstCapacity);
RETURN_ERROR_IF(srcSize + flSize > dstCapacity, dstSize_tooSmall, "");
switch(flSize)
{
case 1: /* 2 - 1 - 5 */
ostart[0] = (BYTE)((U32)set_basic + (srcSize<<3));
break;
case 2: /* 2 - 2 - 12 */
MEM_writeLE16(ostart, (U16)((U32)set_basic + (1<<2) + (srcSize<<4)));
break;
case 3: /* 2 - 2 - 20 */
MEM_writeLE32(ostart, (U32)((U32)set_basic + (3<<2) + (srcSize<<4)));
break;
default: /* not necessary : flSize is {1,2,3} */
assert(0);
}
ZSTD_memcpy(ostart + flSize, src, srcSize);
DEBUGLOG(5, "Raw (uncompressed) literals: %u -> %u", (U32)srcSize, (U32)(srcSize + flSize));
return srcSize + flSize;
}
static int allBytesIdentical(const void* src, size_t srcSize)
{
assert(srcSize >= 1);
assert(src != NULL);
{ const BYTE b = ((const BYTE*)src)[0];
size_t p;
for (p=1; p<srcSize; p++) {
if (((const BYTE*)src)[p] != b) return 0;
}
return 1;
}
}
size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize)
{
BYTE* const ostart = (BYTE*)dst;
U32 const flSize = 1 + (srcSize>31) + (srcSize>4095);
assert(dstCapacity >= 4); (void)dstCapacity;
assert(allBytesIdentical(src, srcSize));
switch(flSize)
{
case 1: /* 2 - 1 - 5 */
ostart[0] = (BYTE)((U32)set_rle + (srcSize<<3));
break;
case 2: /* 2 - 2 - 12 */
MEM_writeLE16(ostart, (U16)((U32)set_rle + (1<<2) + (srcSize<<4)));
break;
case 3: /* 2 - 2 - 20 */
MEM_writeLE32(ostart, (U32)((U32)set_rle + (3<<2) + (srcSize<<4)));
break;
default: /* not necessary : flSize is {1,2,3} */
assert(0);
}
ostart[flSize] = *(const BYTE*)src;
DEBUGLOG(5, "RLE : Repeated Literal (%02X: %u times) -> %u bytes encoded", ((const BYTE*)src)[0], (U32)srcSize, (U32)flSize + 1);
return flSize+1;
}
/* ZSTD_minLiteralsToCompress() :
* returns minimal amount of literals
* for literal compression to even be attempted.
* Minimum is made tighter as compression strategy increases.
*/
static size_t
ZSTD_minLiteralsToCompress(ZSTD_strategy strategy, HUF_repeat huf_repeat)
{
assert((int)strategy >= 0);
assert((int)strategy <= 9);
/* btultra2 : min 8 bytes;
* then 2x larger for each successive compression strategy
* max threshold 64 bytes */
{ int const shift = MIN(9-(int)strategy, 3);
size_t const mintc = (huf_repeat == HUF_repeat_valid) ? 6 : (size_t)8 << shift;
DEBUGLOG(7, "minLiteralsToCompress = %zu", mintc);
return mintc;
}
}
size_t ZSTD_compressLiterals (
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
void* entropyWorkspace, size_t entropyWorkspaceSize,
const ZSTD_hufCTables_t* prevHuf,
ZSTD_hufCTables_t* nextHuf,
ZSTD_strategy strategy,
int disableLiteralCompression,
int suspectUncompressible,
int bmi2)
{
size_t const lhSize = 3 + (srcSize >= 1 KB) + (srcSize >= 16 KB);
BYTE* const ostart = (BYTE*)dst;
U32 singleStream = srcSize < 256;
symbolEncodingType_e hType = set_compressed;
size_t cLitSize;
DEBUGLOG(5,"ZSTD_compressLiterals (disableLiteralCompression=%i, srcSize=%u, dstCapacity=%zu)",
disableLiteralCompression, (U32)srcSize, dstCapacity);
DEBUGLOG(6, "Completed literals listing (%zu bytes)", showHexa(src, srcSize));
/* Prepare nextEntropy assuming reusing the existing table */
ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
if (disableLiteralCompression)
return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
/* if too small, don't even attempt compression (speed opt) */
if (srcSize < ZSTD_minLiteralsToCompress(strategy, prevHuf->repeatMode))
return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
RETURN_ERROR_IF(dstCapacity < lhSize+1, dstSize_tooSmall, "not enough space for compression");
{ HUF_repeat repeat = prevHuf->repeatMode;
int const flags = 0
| (bmi2 ? HUF_flags_bmi2 : 0)
| (strategy < ZSTD_lazy && srcSize <= 1024 ? HUF_flags_preferRepeat : 0)
| (strategy >= HUF_OPTIMAL_DEPTH_THRESHOLD ? HUF_flags_optimalDepth : 0)
| (suspectUncompressible ? HUF_flags_suspectUncompressible : 0);
typedef size_t (*huf_compress_f)(void*, size_t, const void*, size_t, unsigned, unsigned, void*, size_t, HUF_CElt*, HUF_repeat*, int);
huf_compress_f huf_compress;
if (repeat == HUF_repeat_valid && lhSize == 3) singleStream = 1;
huf_compress = singleStream ? HUF_compress1X_repeat : HUF_compress4X_repeat;
cLitSize = huf_compress(ostart+lhSize, dstCapacity-lhSize,
src, srcSize,
HUF_SYMBOLVALUE_MAX, LitHufLog,
entropyWorkspace, entropyWorkspaceSize,
(HUF_CElt*)nextHuf->CTable,
&repeat, flags);
DEBUGLOG(5, "%zu literals compressed into %zu bytes (before header)", srcSize, cLitSize);
if (repeat != HUF_repeat_none) {
/* reused the existing table */
DEBUGLOG(5, "reusing statistics from previous huffman block");
hType = set_repeat;
}
}
{ size_t const minGain = ZSTD_minGain(srcSize, strategy);
if ((cLitSize==0) || (cLitSize >= srcSize - minGain) || ERR_isError(cLitSize)) {
ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize);
} }
if (cLitSize==1) {
/* A return value of 1 signals that the alphabet consists of a single symbol.
* However, in some rare circumstances, it could be the compressed size (a single byte).
* For that outcome to have a chance to happen, it's necessary that `srcSize < 8`.
* (it's also necessary to not generate statistics).
* Therefore, in such a case, actively check that all bytes are identical. */
if ((srcSize >= 8) || allBytesIdentical(src, srcSize)) {
ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf));
return ZSTD_compressRleLiteralsBlock(dst, dstCapacity, src, srcSize);
} }
if (hType == set_compressed) {
/* using a newly constructed table */
nextHuf->repeatMode = HUF_repeat_check;
}
/* Build header */
switch(lhSize)
{
case 3: /* 2 - 2 - 10 - 10 */
if (!singleStream) assert(srcSize >= MIN_LITERALS_FOR_4_STREAMS);
{ U32 const lhc = hType + ((U32)(!singleStream) << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<14);
MEM_writeLE24(ostart, lhc);
break;
}
case 4: /* 2 - 2 - 14 - 14 */
assert(srcSize >= MIN_LITERALS_FOR_4_STREAMS);
{ U32 const lhc = hType + (2 << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<18);
MEM_writeLE32(ostart, lhc);
break;
}
case 5: /* 2 - 2 - 18 - 18 */
assert(srcSize >= MIN_LITERALS_FOR_4_STREAMS);
{ U32 const lhc = hType + (3 << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<22);
MEM_writeLE32(ostart, lhc);
ostart[4] = (BYTE)(cLitSize >> 10);
break;
}
default: /* not possible : lhSize is {3,4,5} */
assert(0);
}
DEBUGLOG(5, "Compressed literals: %u -> %u", (U32)srcSize, (U32)(lhSize+cLitSize));
return lhSize+cLitSize;
}

39
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_compress_literals.h vendored Normal file
View file

@ -0,0 +1,39 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMPRESS_LITERALS_H
#define ZSTD_COMPRESS_LITERALS_H
#include "zstd_compress_internal.h" /* ZSTD_hufCTables_t, ZSTD_minGain() */
size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/* ZSTD_compressRleLiteralsBlock() :
* Conditions :
* - All bytes in @src are identical
* - dstCapacity >= 4 */
size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/* ZSTD_compressLiterals():
* @entropyWorkspace: must be aligned on 4-bytes boundaries
* @entropyWorkspaceSize : must be >= HUF_WORKSPACE_SIZE
* @suspectUncompressible: sampling checks, to potentially skip huffman coding
*/
size_t ZSTD_compressLiterals (void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
void* entropyWorkspace, size_t entropyWorkspaceSize,
const ZSTD_hufCTables_t* prevHuf,
ZSTD_hufCTables_t* nextHuf,
ZSTD_strategy strategy, int disableLiteralCompression,
int suspectUncompressible,
int bmi2);
#endif /* ZSTD_COMPRESS_LITERALS_H */

442
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_compress_sequences.c vendored Normal file
View file

@ -0,0 +1,442 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*-*************************************
* Dependencies
***************************************/
#include "zstd_compress_sequences.h"
/**
* -log2(x / 256) lookup table for x in [0, 256).
* If x == 0: Return 0
* Else: Return floor(-log2(x / 256) * 256)
*/
static unsigned const kInverseProbabilityLog256[256] = {
0, 2048, 1792, 1642, 1536, 1453, 1386, 1329, 1280, 1236, 1197, 1162,
1130, 1100, 1073, 1047, 1024, 1001, 980, 960, 941, 923, 906, 889,
874, 859, 844, 830, 817, 804, 791, 779, 768, 756, 745, 734,
724, 714, 704, 694, 685, 676, 667, 658, 650, 642, 633, 626,
618, 610, 603, 595, 588, 581, 574, 567, 561, 554, 548, 542,
535, 529, 523, 517, 512, 506, 500, 495, 489, 484, 478, 473,
468, 463, 458, 453, 448, 443, 438, 434, 429, 424, 420, 415,
411, 407, 402, 398, 394, 390, 386, 382, 377, 373, 370, 366,
362, 358, 354, 350, 347, 343, 339, 336, 332, 329, 325, 322,
318, 315, 311, 308, 305, 302, 298, 295, 292, 289, 286, 282,
279, 276, 273, 270, 267, 264, 261, 258, 256, 253, 250, 247,
244, 241, 239, 236, 233, 230, 228, 225, 222, 220, 217, 215,
212, 209, 207, 204, 202, 199, 197, 194, 192, 190, 187, 185,
182, 180, 178, 175, 173, 171, 168, 166, 164, 162, 159, 157,
155, 153, 151, 149, 146, 144, 142, 140, 138, 136, 134, 132,
130, 128, 126, 123, 121, 119, 117, 115, 114, 112, 110, 108,
106, 104, 102, 100, 98, 96, 94, 93, 91, 89, 87, 85,
83, 82, 80, 78, 76, 74, 73, 71, 69, 67, 66, 64,
62, 61, 59, 57, 55, 54, 52, 50, 49, 47, 46, 44,
42, 41, 39, 37, 36, 34, 33, 31, 30, 28, 26, 25,
23, 22, 20, 19, 17, 16, 14, 13, 11, 10, 8, 7,
5, 4, 2, 1,
};
static unsigned ZSTD_getFSEMaxSymbolValue(FSE_CTable const* ctable) {
void const* ptr = ctable;
U16 const* u16ptr = (U16 const*)ptr;
U32 const maxSymbolValue = MEM_read16(u16ptr + 1);
return maxSymbolValue;
}
/**
* Returns true if we should use ncount=-1 else we should
* use ncount=1 for low probability symbols instead.
*/
static unsigned ZSTD_useLowProbCount(size_t const nbSeq)
{
/* Heuristic: This should cover most blocks <= 16K and
* start to fade out after 16K to about 32K depending on
* compressibility.
*/
return nbSeq >= 2048;
}
/**
* Returns the cost in bytes of encoding the normalized count header.
* Returns an error if any of the helper functions return an error.
*/
static size_t ZSTD_NCountCost(unsigned const* count, unsigned const max,
size_t const nbSeq, unsigned const FSELog)
{
BYTE wksp[FSE_NCOUNTBOUND];
S16 norm[MaxSeq + 1];
const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max);
FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq, max, ZSTD_useLowProbCount(nbSeq)), "");
return FSE_writeNCount(wksp, sizeof(wksp), norm, max, tableLog);
}
/**
* Returns the cost in bits of encoding the distribution described by count
* using the entropy bound.
*/
static size_t ZSTD_entropyCost(unsigned const* count, unsigned const max, size_t const total)
{
unsigned cost = 0;
unsigned s;
assert(total > 0);
for (s = 0; s <= max; ++s) {
unsigned norm = (unsigned)((256 * count[s]) / total);
if (count[s] != 0 && norm == 0)
norm = 1;
assert(count[s] < total);
cost += count[s] * kInverseProbabilityLog256[norm];
}
return cost >> 8;
}
/**
* Returns the cost in bits of encoding the distribution in count using ctable.
* Returns an error if ctable cannot represent all the symbols in count.
*/
size_t ZSTD_fseBitCost(
FSE_CTable const* ctable,
unsigned const* count,
unsigned const max)
{
unsigned const kAccuracyLog = 8;
size_t cost = 0;
unsigned s;
FSE_CState_t cstate;
FSE_initCState(&cstate, ctable);
if (ZSTD_getFSEMaxSymbolValue(ctable) < max) {
DEBUGLOG(5, "Repeat FSE_CTable has maxSymbolValue %u < %u",
ZSTD_getFSEMaxSymbolValue(ctable), max);
return ERROR(GENERIC);
}
for (s = 0; s <= max; ++s) {
unsigned const tableLog = cstate.stateLog;
unsigned const badCost = (tableLog + 1) << kAccuracyLog;
unsigned const bitCost = FSE_bitCost(cstate.symbolTT, tableLog, s, kAccuracyLog);
if (count[s] == 0)
continue;
if (bitCost >= badCost) {
DEBUGLOG(5, "Repeat FSE_CTable has Prob[%u] == 0", s);
return ERROR(GENERIC);
}
cost += (size_t)count[s] * bitCost;
}
return cost >> kAccuracyLog;
}
/**
* Returns the cost in bits of encoding the distribution in count using the
* table described by norm. The max symbol support by norm is assumed >= max.
* norm must be valid for every symbol with non-zero probability in count.
*/
size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog,
unsigned const* count, unsigned const max)
{
unsigned const shift = 8 - accuracyLog;
size_t cost = 0;
unsigned s;
assert(accuracyLog <= 8);
for (s = 0; s <= max; ++s) {
unsigned const normAcc = (norm[s] != -1) ? (unsigned)norm[s] : 1;
unsigned const norm256 = normAcc << shift;
assert(norm256 > 0);
assert(norm256 < 256);
cost += count[s] * kInverseProbabilityLog256[norm256];
}
return cost >> 8;
}
symbolEncodingType_e
ZSTD_selectEncodingType(
FSE_repeat* repeatMode, unsigned const* count, unsigned const max,
size_t const mostFrequent, size_t nbSeq, unsigned const FSELog,
FSE_CTable const* prevCTable,
short const* defaultNorm, U32 defaultNormLog,
ZSTD_defaultPolicy_e const isDefaultAllowed,
ZSTD_strategy const strategy)
{
ZSTD_STATIC_ASSERT(ZSTD_defaultDisallowed == 0 && ZSTD_defaultAllowed != 0);
if (mostFrequent == nbSeq) {
*repeatMode = FSE_repeat_none;
if (isDefaultAllowed && nbSeq <= 2) {
/* Prefer set_basic over set_rle when there are 2 or fewer symbols,
* since RLE uses 1 byte, but set_basic uses 5-6 bits per symbol.
* If basic encoding isn't possible, always choose RLE.
*/
DEBUGLOG(5, "Selected set_basic");
return set_basic;
}
DEBUGLOG(5, "Selected set_rle");
return set_rle;
}
if (strategy < ZSTD_lazy) {
if (isDefaultAllowed) {
size_t const staticFse_nbSeq_max = 1000;
size_t const mult = 10 - strategy;
size_t const baseLog = 3;
size_t const dynamicFse_nbSeq_min = (((size_t)1 << defaultNormLog) * mult) >> baseLog; /* 28-36 for offset, 56-72 for lengths */
assert(defaultNormLog >= 5 && defaultNormLog <= 6); /* xx_DEFAULTNORMLOG */
assert(mult <= 9 && mult >= 7);
if ( (*repeatMode == FSE_repeat_valid)
&& (nbSeq < staticFse_nbSeq_max) ) {
DEBUGLOG(5, "Selected set_repeat");
return set_repeat;
}
if ( (nbSeq < dynamicFse_nbSeq_min)
|| (mostFrequent < (nbSeq >> (defaultNormLog-1))) ) {
DEBUGLOG(5, "Selected set_basic");
/* The format allows default tables to be repeated, but it isn't useful.
* When using simple heuristics to select encoding type, we don't want
* to confuse these tables with dictionaries. When running more careful
* analysis, we don't need to waste time checking both repeating tables
* and default tables.
*/
*repeatMode = FSE_repeat_none;
return set_basic;
}
}
} else {
size_t const basicCost = isDefaultAllowed ? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, count, max) : ERROR(GENERIC);
size_t const repeatCost = *repeatMode != FSE_repeat_none ? ZSTD_fseBitCost(prevCTable, count, max) : ERROR(GENERIC);
size_t const NCountCost = ZSTD_NCountCost(count, max, nbSeq, FSELog);
size_t const compressedCost = (NCountCost << 3) + ZSTD_entropyCost(count, max, nbSeq);
if (isDefaultAllowed) {
assert(!ZSTD_isError(basicCost));
assert(!(*repeatMode == FSE_repeat_valid && ZSTD_isError(repeatCost)));
}
assert(!ZSTD_isError(NCountCost));
assert(compressedCost < ERROR(maxCode));
DEBUGLOG(5, "Estimated bit costs: basic=%u\trepeat=%u\tcompressed=%u",
(unsigned)basicCost, (unsigned)repeatCost, (unsigned)compressedCost);
if (basicCost <= repeatCost && basicCost <= compressedCost) {
DEBUGLOG(5, "Selected set_basic");
assert(isDefaultAllowed);
*repeatMode = FSE_repeat_none;
return set_basic;
}
if (repeatCost <= compressedCost) {
DEBUGLOG(5, "Selected set_repeat");
assert(!ZSTD_isError(repeatCost));
return set_repeat;
}
assert(compressedCost < basicCost && compressedCost < repeatCost);
}
DEBUGLOG(5, "Selected set_compressed");
*repeatMode = FSE_repeat_check;
return set_compressed;
}
typedef struct {
S16 norm[MaxSeq + 1];
U32 wksp[FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(MaxSeq, MaxFSELog)];
} ZSTD_BuildCTableWksp;
size_t
ZSTD_buildCTable(void* dst, size_t dstCapacity,
FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type,
unsigned* count, U32 max,
const BYTE* codeTable, size_t nbSeq,
const S16* defaultNorm, U32 defaultNormLog, U32 defaultMax,
const FSE_CTable* prevCTable, size_t prevCTableSize,
void* entropyWorkspace, size_t entropyWorkspaceSize)
{
BYTE* op = (BYTE*)dst;
const BYTE* const oend = op + dstCapacity;
DEBUGLOG(6, "ZSTD_buildCTable (dstCapacity=%u)", (unsigned)dstCapacity);
switch (type) {
case set_rle:
FORWARD_IF_ERROR(FSE_buildCTable_rle(nextCTable, (BYTE)max), "");
RETURN_ERROR_IF(dstCapacity==0, dstSize_tooSmall, "not enough space");
*op = codeTable[0];
return 1;
case set_repeat:
ZSTD_memcpy(nextCTable, prevCTable, prevCTableSize);
return 0;
case set_basic:
FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, entropyWorkspace, entropyWorkspaceSize), ""); /* note : could be pre-calculated */
return 0;
case set_compressed: {
ZSTD_BuildCTableWksp* wksp = (ZSTD_BuildCTableWksp*)entropyWorkspace;
size_t nbSeq_1 = nbSeq;
const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max);
if (count[codeTable[nbSeq-1]] > 1) {
count[codeTable[nbSeq-1]]--;
nbSeq_1--;
}
assert(nbSeq_1 > 1);
assert(entropyWorkspaceSize >= sizeof(ZSTD_BuildCTableWksp));
(void)entropyWorkspaceSize;
FORWARD_IF_ERROR(FSE_normalizeCount(wksp->norm, tableLog, count, nbSeq_1, max, ZSTD_useLowProbCount(nbSeq_1)), "FSE_normalizeCount failed");
assert(oend >= op);
{ size_t const NCountSize = FSE_writeNCount(op, (size_t)(oend - op), wksp->norm, max, tableLog); /* overflow protected */
FORWARD_IF_ERROR(NCountSize, "FSE_writeNCount failed");
FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, wksp->norm, max, tableLog, wksp->wksp, sizeof(wksp->wksp)), "FSE_buildCTable_wksp failed");
return NCountSize;
}
}
default: assert(0); RETURN_ERROR(GENERIC, "impossible to reach");
}
}
FORCE_INLINE_TEMPLATE size_t
ZSTD_encodeSequences_body(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets)
{
BIT_CStream_t blockStream;
FSE_CState_t stateMatchLength;
FSE_CState_t stateOffsetBits;
FSE_CState_t stateLitLength;
RETURN_ERROR_IF(
ERR_isError(BIT_initCStream(&blockStream, dst, dstCapacity)),
dstSize_tooSmall, "not enough space remaining");
DEBUGLOG(6, "available space for bitstream : %i (dstCapacity=%u)",
(int)(blockStream.endPtr - blockStream.startPtr),
(unsigned)dstCapacity);
/* first symbols */
FSE_initCState2(&stateMatchLength, CTable_MatchLength, mlCodeTable[nbSeq-1]);
FSE_initCState2(&stateOffsetBits, CTable_OffsetBits, ofCodeTable[nbSeq-1]);
FSE_initCState2(&stateLitLength, CTable_LitLength, llCodeTable[nbSeq-1]);
BIT_addBits(&blockStream, sequences[nbSeq-1].litLength, LL_bits[llCodeTable[nbSeq-1]]);
if (MEM_32bits()) BIT_flushBits(&blockStream);
BIT_addBits(&blockStream, sequences[nbSeq-1].mlBase, ML_bits[mlCodeTable[nbSeq-1]]);
if (MEM_32bits()) BIT_flushBits(&blockStream);
if (longOffsets) {
U32 const ofBits = ofCodeTable[nbSeq-1];
unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
if (extraBits) {
BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, extraBits);
BIT_flushBits(&blockStream);
}
BIT_addBits(&blockStream, sequences[nbSeq-1].offBase >> extraBits,
ofBits - extraBits);
} else {
BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, ofCodeTable[nbSeq-1]);
}
BIT_flushBits(&blockStream);
{ size_t n;
for (n=nbSeq-2 ; n<nbSeq ; n--) { /* intentional underflow */
BYTE const llCode = llCodeTable[n];
BYTE const ofCode = ofCodeTable[n];
BYTE const mlCode = mlCodeTable[n];
U32 const llBits = LL_bits[llCode];
U32 const ofBits = ofCode;
U32 const mlBits = ML_bits[mlCode];
DEBUGLOG(6, "encoding: litlen:%2u - matchlen:%2u - offCode:%7u",
(unsigned)sequences[n].litLength,
(unsigned)sequences[n].mlBase + MINMATCH,
(unsigned)sequences[n].offBase);
/* 32b*/ /* 64b*/
/* (7)*/ /* (7)*/
FSE_encodeSymbol(&blockStream, &stateOffsetBits, ofCode); /* 15 */ /* 15 */
FSE_encodeSymbol(&blockStream, &stateMatchLength, mlCode); /* 24 */ /* 24 */
if (MEM_32bits()) BIT_flushBits(&blockStream); /* (7)*/
FSE_encodeSymbol(&blockStream, &stateLitLength, llCode); /* 16 */ /* 33 */
if (MEM_32bits() || (ofBits+mlBits+llBits >= 64-7-(LLFSELog+MLFSELog+OffFSELog)))
BIT_flushBits(&blockStream); /* (7)*/
BIT_addBits(&blockStream, sequences[n].litLength, llBits);
if (MEM_32bits() && ((llBits+mlBits)>24)) BIT_flushBits(&blockStream);
BIT_addBits(&blockStream, sequences[n].mlBase, mlBits);
if (MEM_32bits() || (ofBits+mlBits+llBits > 56)) BIT_flushBits(&blockStream);
if (longOffsets) {
unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
if (extraBits) {
BIT_addBits(&blockStream, sequences[n].offBase, extraBits);
BIT_flushBits(&blockStream); /* (7)*/
}
BIT_addBits(&blockStream, sequences[n].offBase >> extraBits,
ofBits - extraBits); /* 31 */
} else {
BIT_addBits(&blockStream, sequences[n].offBase, ofBits); /* 31 */
}
BIT_flushBits(&blockStream); /* (7)*/
DEBUGLOG(7, "remaining space : %i", (int)(blockStream.endPtr - blockStream.ptr));
} }
DEBUGLOG(6, "ZSTD_encodeSequences: flushing ML state with %u bits", stateMatchLength.stateLog);
FSE_flushCState(&blockStream, &stateMatchLength);
DEBUGLOG(6, "ZSTD_encodeSequences: flushing Off state with %u bits", stateOffsetBits.stateLog);
FSE_flushCState(&blockStream, &stateOffsetBits);
DEBUGLOG(6, "ZSTD_encodeSequences: flushing LL state with %u bits", stateLitLength.stateLog);
FSE_flushCState(&blockStream, &stateLitLength);
{ size_t const streamSize = BIT_closeCStream(&blockStream);
RETURN_ERROR_IF(streamSize==0, dstSize_tooSmall, "not enough space");
return streamSize;
}
}
static size_t
ZSTD_encodeSequences_default(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets)
{
return ZSTD_encodeSequences_body(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
CTable_OffsetBits, ofCodeTable,
CTable_LitLength, llCodeTable,
sequences, nbSeq, longOffsets);
}
#if DYNAMIC_BMI2
static BMI2_TARGET_ATTRIBUTE size_t
ZSTD_encodeSequences_bmi2(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets)
{
return ZSTD_encodeSequences_body(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
CTable_OffsetBits, ofCodeTable,
CTable_LitLength, llCodeTable,
sequences, nbSeq, longOffsets);
}
#endif
size_t ZSTD_encodeSequences(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2)
{
DEBUGLOG(5, "ZSTD_encodeSequences: dstCapacity = %u", (unsigned)dstCapacity);
#if DYNAMIC_BMI2
if (bmi2) {
return ZSTD_encodeSequences_bmi2(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
CTable_OffsetBits, ofCodeTable,
CTable_LitLength, llCodeTable,
sequences, nbSeq, longOffsets);
}
#endif
(void)bmi2;
return ZSTD_encodeSequences_default(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
CTable_OffsetBits, ofCodeTable,
CTable_LitLength, llCodeTable,
sequences, nbSeq, longOffsets);
}

54
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_compress_sequences.h vendored Normal file
View file

@ -0,0 +1,54 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMPRESS_SEQUENCES_H
#define ZSTD_COMPRESS_SEQUENCES_H
#include "../common/fse.h" /* FSE_repeat, FSE_CTable */
#include "../common/zstd_internal.h" /* symbolEncodingType_e, ZSTD_strategy */
typedef enum {
ZSTD_defaultDisallowed = 0,
ZSTD_defaultAllowed = 1
} ZSTD_defaultPolicy_e;
symbolEncodingType_e
ZSTD_selectEncodingType(
FSE_repeat* repeatMode, unsigned const* count, unsigned const max,
size_t const mostFrequent, size_t nbSeq, unsigned const FSELog,
FSE_CTable const* prevCTable,
short const* defaultNorm, U32 defaultNormLog,
ZSTD_defaultPolicy_e const isDefaultAllowed,
ZSTD_strategy const strategy);
size_t
ZSTD_buildCTable(void* dst, size_t dstCapacity,
FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type,
unsigned* count, U32 max,
const BYTE* codeTable, size_t nbSeq,
const S16* defaultNorm, U32 defaultNormLog, U32 defaultMax,
const FSE_CTable* prevCTable, size_t prevCTableSize,
void* entropyWorkspace, size_t entropyWorkspaceSize);
size_t ZSTD_encodeSequences(
void* dst, size_t dstCapacity,
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2);
size_t ZSTD_fseBitCost(
FSE_CTable const* ctable,
unsigned const* count,
unsigned const max);
size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog,
unsigned const* count, unsigned const max);
#endif /* ZSTD_COMPRESS_SEQUENCES_H */

577
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_compress_superblock.c vendored Normal file
View file

@ -0,0 +1,577 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*-*************************************
* Dependencies
***************************************/
#include "zstd_compress_superblock.h"
#include "../common/zstd_internal.h" /* ZSTD_getSequenceLength */
#include "hist.h" /* HIST_countFast_wksp */
#include "zstd_compress_internal.h" /* ZSTD_[huf|fse|entropy]CTablesMetadata_t */
#include "zstd_compress_sequences.h"
#include "zstd_compress_literals.h"
/** ZSTD_compressSubBlock_literal() :
* Compresses literals section for a sub-block.
* When we have to write the Huffman table we will sometimes choose a header
* size larger than necessary. This is because we have to pick the header size
* before we know the table size + compressed size, so we have a bound on the
* table size. If we guessed incorrectly, we fall back to uncompressed literals.
*
* We write the header when writeEntropy=1 and set entropyWritten=1 when we succeeded
* in writing the header, otherwise it is set to 0.
*
* hufMetadata->hType has literals block type info.
* If it is set_basic, all sub-blocks literals section will be Raw_Literals_Block.
* If it is set_rle, all sub-blocks literals section will be RLE_Literals_Block.
* If it is set_compressed, first sub-block's literals section will be Compressed_Literals_Block
* If it is set_compressed, first sub-block's literals section will be Treeless_Literals_Block
* and the following sub-blocks' literals sections will be Treeless_Literals_Block.
* @return : compressed size of literals section of a sub-block
* Or 0 if unable to compress.
* Or error code */
static size_t
ZSTD_compressSubBlock_literal(const HUF_CElt* hufTable,
const ZSTD_hufCTablesMetadata_t* hufMetadata,
const BYTE* literals, size_t litSize,
void* dst, size_t dstSize,
const int bmi2, int writeEntropy, int* entropyWritten)
{
size_t const header = writeEntropy ? 200 : 0;
size_t const lhSize = 3 + (litSize >= (1 KB - header)) + (litSize >= (16 KB - header));
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstSize;
BYTE* op = ostart + lhSize;
U32 const singleStream = lhSize == 3;
symbolEncodingType_e hType = writeEntropy ? hufMetadata->hType : set_repeat;
size_t cLitSize = 0;
DEBUGLOG(5, "ZSTD_compressSubBlock_literal (litSize=%zu, lhSize=%zu, writeEntropy=%d)", litSize, lhSize, writeEntropy);
*entropyWritten = 0;
if (litSize == 0 || hufMetadata->hType == set_basic) {
DEBUGLOG(5, "ZSTD_compressSubBlock_literal using raw literal");
return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize);
} else if (hufMetadata->hType == set_rle) {
DEBUGLOG(5, "ZSTD_compressSubBlock_literal using rle literal");
return ZSTD_compressRleLiteralsBlock(dst, dstSize, literals, litSize);
}
assert(litSize > 0);
assert(hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat);
if (writeEntropy && hufMetadata->hType == set_compressed) {
ZSTD_memcpy(op, hufMetadata->hufDesBuffer, hufMetadata->hufDesSize);
op += hufMetadata->hufDesSize;
cLitSize += hufMetadata->hufDesSize;
DEBUGLOG(5, "ZSTD_compressSubBlock_literal (hSize=%zu)", hufMetadata->hufDesSize);
}
{ int const flags = bmi2 ? HUF_flags_bmi2 : 0;
const size_t cSize = singleStream ? HUF_compress1X_usingCTable(op, oend-op, literals, litSize, hufTable, flags)
: HUF_compress4X_usingCTable(op, oend-op, literals, litSize, hufTable, flags);
op += cSize;
cLitSize += cSize;
if (cSize == 0 || ERR_isError(cSize)) {
DEBUGLOG(5, "Failed to write entropy tables %s", ZSTD_getErrorName(cSize));
return 0;
}
/* If we expand and we aren't writing a header then emit uncompressed */
if (!writeEntropy && cLitSize >= litSize) {
DEBUGLOG(5, "ZSTD_compressSubBlock_literal using raw literal because uncompressible");
return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize);
}
/* If we are writing headers then allow expansion that doesn't change our header size. */
if (lhSize < (size_t)(3 + (cLitSize >= 1 KB) + (cLitSize >= 16 KB))) {
assert(cLitSize > litSize);
DEBUGLOG(5, "Literals expanded beyond allowed header size");
return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize);
}
DEBUGLOG(5, "ZSTD_compressSubBlock_literal (cSize=%zu)", cSize);
}
/* Build header */
switch(lhSize)
{
case 3: /* 2 - 2 - 10 - 10 */
{ U32 const lhc = hType + ((!singleStream) << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<14);
MEM_writeLE24(ostart, lhc);
break;
}
case 4: /* 2 - 2 - 14 - 14 */
{ U32 const lhc = hType + (2 << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<18);
MEM_writeLE32(ostart, lhc);
break;
}
case 5: /* 2 - 2 - 18 - 18 */
{ U32 const lhc = hType + (3 << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<22);
MEM_writeLE32(ostart, lhc);
ostart[4] = (BYTE)(cLitSize >> 10);
break;
}
default: /* not possible : lhSize is {3,4,5} */
assert(0);
}
*entropyWritten = 1;
DEBUGLOG(5, "Compressed literals: %u -> %u", (U32)litSize, (U32)(op-ostart));
return op-ostart;
}
static size_t
ZSTD_seqDecompressedSize(seqStore_t const* seqStore,
const seqDef* sequences, size_t nbSeq,
size_t litSize, int lastSequence)
{
const seqDef* const sstart = sequences;
const seqDef* const send = sequences + nbSeq;
const seqDef* sp = sstart;
size_t matchLengthSum = 0;
size_t litLengthSum = 0;
(void)(litLengthSum); /* suppress unused variable warning on some environments */
while (send-sp > 0) {
ZSTD_sequenceLength const seqLen = ZSTD_getSequenceLength(seqStore, sp);
litLengthSum += seqLen.litLength;
matchLengthSum += seqLen.matchLength;
sp++;
}
assert(litLengthSum <= litSize);
if (!lastSequence) {
assert(litLengthSum == litSize);
}
return matchLengthSum + litSize;
}
/** ZSTD_compressSubBlock_sequences() :
* Compresses sequences section for a sub-block.
* fseMetadata->llType, fseMetadata->ofType, and fseMetadata->mlType have
* symbol compression modes for the super-block.
* The first successfully compressed block will have these in its header.
* We set entropyWritten=1 when we succeed in compressing the sequences.
* The following sub-blocks will always have repeat mode.
* @return : compressed size of sequences section of a sub-block
* Or 0 if it is unable to compress
* Or error code. */
static size_t
ZSTD_compressSubBlock_sequences(const ZSTD_fseCTables_t* fseTables,
const ZSTD_fseCTablesMetadata_t* fseMetadata,
const seqDef* sequences, size_t nbSeq,
const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode,
const ZSTD_CCtx_params* cctxParams,
void* dst, size_t dstCapacity,
const int bmi2, int writeEntropy, int* entropyWritten)
{
const int longOffsets = cctxParams->cParams.windowLog > STREAM_ACCUMULATOR_MIN;
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstCapacity;
BYTE* op = ostart;
BYTE* seqHead;
DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (nbSeq=%zu, writeEntropy=%d, longOffsets=%d)", nbSeq, writeEntropy, longOffsets);
*entropyWritten = 0;
/* Sequences Header */
RETURN_ERROR_IF((oend-op) < 3 /*max nbSeq Size*/ + 1 /*seqHead*/,
dstSize_tooSmall, "");
if (nbSeq < 0x7F)
*op++ = (BYTE)nbSeq;
else if (nbSeq < LONGNBSEQ)
op[0] = (BYTE)((nbSeq>>8) + 0x80), op[1] = (BYTE)nbSeq, op+=2;
else
op[0]=0xFF, MEM_writeLE16(op+1, (U16)(nbSeq - LONGNBSEQ)), op+=3;
if (nbSeq==0) {
return op - ostart;
}
/* seqHead : flags for FSE encoding type */
seqHead = op++;
DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (seqHeadSize=%u)", (unsigned)(op-ostart));
if (writeEntropy) {
const U32 LLtype = fseMetadata->llType;
const U32 Offtype = fseMetadata->ofType;
const U32 MLtype = fseMetadata->mlType;
DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (fseTablesSize=%zu)", fseMetadata->fseTablesSize);
*seqHead = (BYTE)((LLtype<<6) + (Offtype<<4) + (MLtype<<2));
ZSTD_memcpy(op, fseMetadata->fseTablesBuffer, fseMetadata->fseTablesSize);
op += fseMetadata->fseTablesSize;
} else {
const U32 repeat = set_repeat;
*seqHead = (BYTE)((repeat<<6) + (repeat<<4) + (repeat<<2));
}
{ size_t const bitstreamSize = ZSTD_encodeSequences(
op, oend - op,
fseTables->matchlengthCTable, mlCode,
fseTables->offcodeCTable, ofCode,
fseTables->litlengthCTable, llCode,
sequences, nbSeq,
longOffsets, bmi2);
FORWARD_IF_ERROR(bitstreamSize, "ZSTD_encodeSequences failed");
op += bitstreamSize;
/* zstd versions <= 1.3.4 mistakenly report corruption when
* FSE_readNCount() receives a buffer < 4 bytes.
* Fixed by https://github.com/facebook/zstd/pull/1146.
* This can happen when the last set_compressed table present is 2
* bytes and the bitstream is only one byte.
* In this exceedingly rare case, we will simply emit an uncompressed
* block, since it isn't worth optimizing.
*/
#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
if (writeEntropy && fseMetadata->lastCountSize && fseMetadata->lastCountSize + bitstreamSize < 4) {
/* NCountSize >= 2 && bitstreamSize > 0 ==> lastCountSize == 3 */
assert(fseMetadata->lastCountSize + bitstreamSize == 3);
DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.3.4 by "
"emitting an uncompressed block.");
return 0;
}
#endif
DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (bitstreamSize=%zu)", bitstreamSize);
}
/* zstd versions <= 1.4.0 mistakenly report error when
* sequences section body size is less than 3 bytes.
* Fixed by https://github.com/facebook/zstd/pull/1664.
* This can happen when the previous sequences section block is compressed
* with rle mode and the current block's sequences section is compressed
* with repeat mode where sequences section body size can be 1 byte.
*/
#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
if (op-seqHead < 4) {
DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.4.0 by emitting "
"an uncompressed block when sequences are < 4 bytes");
return 0;
}
#endif
*entropyWritten = 1;
return op - ostart;
}
/** ZSTD_compressSubBlock() :
* Compresses a single sub-block.
* @return : compressed size of the sub-block
* Or 0 if it failed to compress. */
static size_t ZSTD_compressSubBlock(const ZSTD_entropyCTables_t* entropy,
const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
const seqDef* sequences, size_t nbSeq,
const BYTE* literals, size_t litSize,
const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode,
const ZSTD_CCtx_params* cctxParams,
void* dst, size_t dstCapacity,
const int bmi2,
int writeLitEntropy, int writeSeqEntropy,
int* litEntropyWritten, int* seqEntropyWritten,
U32 lastBlock)
{
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstCapacity;
BYTE* op = ostart + ZSTD_blockHeaderSize;
DEBUGLOG(5, "ZSTD_compressSubBlock (litSize=%zu, nbSeq=%zu, writeLitEntropy=%d, writeSeqEntropy=%d, lastBlock=%d)",
litSize, nbSeq, writeLitEntropy, writeSeqEntropy, lastBlock);
{ size_t cLitSize = ZSTD_compressSubBlock_literal((const HUF_CElt*)entropy->huf.CTable,
&entropyMetadata->hufMetadata, literals, litSize,
op, oend-op, bmi2, writeLitEntropy, litEntropyWritten);
FORWARD_IF_ERROR(cLitSize, "ZSTD_compressSubBlock_literal failed");
if (cLitSize == 0) return 0;
op += cLitSize;
}
{ size_t cSeqSize = ZSTD_compressSubBlock_sequences(&entropy->fse,
&entropyMetadata->fseMetadata,
sequences, nbSeq,
llCode, mlCode, ofCode,
cctxParams,
op, oend-op,
bmi2, writeSeqEntropy, seqEntropyWritten);
FORWARD_IF_ERROR(cSeqSize, "ZSTD_compressSubBlock_sequences failed");
if (cSeqSize == 0) return 0;
op += cSeqSize;
}
/* Write block header */
{ size_t cSize = (op-ostart)-ZSTD_blockHeaderSize;
U32 const cBlockHeader24 = lastBlock + (((U32)bt_compressed)<<1) + (U32)(cSize << 3);
MEM_writeLE24(ostart, cBlockHeader24);
}
return op-ostart;
}
static size_t ZSTD_estimateSubBlockSize_literal(const BYTE* literals, size_t litSize,
const ZSTD_hufCTables_t* huf,
const ZSTD_hufCTablesMetadata_t* hufMetadata,
void* workspace, size_t wkspSize,
int writeEntropy)
{
unsigned* const countWksp = (unsigned*)workspace;
unsigned maxSymbolValue = 255;
size_t literalSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */
if (hufMetadata->hType == set_basic) return litSize;
else if (hufMetadata->hType == set_rle) return 1;
else if (hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat) {
size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)literals, litSize, workspace, wkspSize);
if (ZSTD_isError(largest)) return litSize;
{ size_t cLitSizeEstimate = HUF_estimateCompressedSize((const HUF_CElt*)huf->CTable, countWksp, maxSymbolValue);
if (writeEntropy) cLitSizeEstimate += hufMetadata->hufDesSize;
return cLitSizeEstimate + literalSectionHeaderSize;
} }
assert(0); /* impossible */
return 0;
}
static size_t ZSTD_estimateSubBlockSize_symbolType(symbolEncodingType_e type,
const BYTE* codeTable, unsigned maxCode,
size_t nbSeq, const FSE_CTable* fseCTable,
const U8* additionalBits,
short const* defaultNorm, U32 defaultNormLog, U32 defaultMax,
void* workspace, size_t wkspSize)
{
unsigned* const countWksp = (unsigned*)workspace;
const BYTE* ctp = codeTable;
const BYTE* const ctStart = ctp;
const BYTE* const ctEnd = ctStart + nbSeq;
size_t cSymbolTypeSizeEstimateInBits = 0;
unsigned max = maxCode;
HIST_countFast_wksp(countWksp, &max, codeTable, nbSeq, workspace, wkspSize); /* can't fail */
if (type == set_basic) {
/* We selected this encoding type, so it must be valid. */
assert(max <= defaultMax);
cSymbolTypeSizeEstimateInBits = max <= defaultMax
? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, countWksp, max)
: ERROR(GENERIC);
} else if (type == set_rle) {
cSymbolTypeSizeEstimateInBits = 0;
} else if (type == set_compressed || type == set_repeat) {
cSymbolTypeSizeEstimateInBits = ZSTD_fseBitCost(fseCTable, countWksp, max);
}
if (ZSTD_isError(cSymbolTypeSizeEstimateInBits)) return nbSeq * 10;
while (ctp < ctEnd) {
if (additionalBits) cSymbolTypeSizeEstimateInBits += additionalBits[*ctp];
else cSymbolTypeSizeEstimateInBits += *ctp; /* for offset, offset code is also the number of additional bits */
ctp++;
}
return cSymbolTypeSizeEstimateInBits / 8;
}
static size_t ZSTD_estimateSubBlockSize_sequences(const BYTE* ofCodeTable,
const BYTE* llCodeTable,
const BYTE* mlCodeTable,
size_t nbSeq,
const ZSTD_fseCTables_t* fseTables,
const ZSTD_fseCTablesMetadata_t* fseMetadata,
void* workspace, size_t wkspSize,
int writeEntropy)
{
size_t const sequencesSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */
size_t cSeqSizeEstimate = 0;
if (nbSeq == 0) return sequencesSectionHeaderSize;
cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->ofType, ofCodeTable, MaxOff,
nbSeq, fseTables->offcodeCTable, NULL,
OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff,
workspace, wkspSize);
cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->llType, llCodeTable, MaxLL,
nbSeq, fseTables->litlengthCTable, LL_bits,
LL_defaultNorm, LL_defaultNormLog, MaxLL,
workspace, wkspSize);
cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->mlType, mlCodeTable, MaxML,
nbSeq, fseTables->matchlengthCTable, ML_bits,
ML_defaultNorm, ML_defaultNormLog, MaxML,
workspace, wkspSize);
if (writeEntropy) cSeqSizeEstimate += fseMetadata->fseTablesSize;
return cSeqSizeEstimate + sequencesSectionHeaderSize;
}
static size_t ZSTD_estimateSubBlockSize(const BYTE* literals, size_t litSize,
const BYTE* ofCodeTable,
const BYTE* llCodeTable,
const BYTE* mlCodeTable,
size_t nbSeq,
const ZSTD_entropyCTables_t* entropy,
const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
void* workspace, size_t wkspSize,
int writeLitEntropy, int writeSeqEntropy) {
size_t cSizeEstimate = 0;
cSizeEstimate += ZSTD_estimateSubBlockSize_literal(literals, litSize,
&entropy->huf, &entropyMetadata->hufMetadata,
workspace, wkspSize, writeLitEntropy);
cSizeEstimate += ZSTD_estimateSubBlockSize_sequences(ofCodeTable, llCodeTable, mlCodeTable,
nbSeq, &entropy->fse, &entropyMetadata->fseMetadata,
workspace, wkspSize, writeSeqEntropy);
return cSizeEstimate + ZSTD_blockHeaderSize;
}
static int ZSTD_needSequenceEntropyTables(ZSTD_fseCTablesMetadata_t const* fseMetadata)
{
if (fseMetadata->llType == set_compressed || fseMetadata->llType == set_rle)
return 1;
if (fseMetadata->mlType == set_compressed || fseMetadata->mlType == set_rle)
return 1;
if (fseMetadata->ofType == set_compressed || fseMetadata->ofType == set_rle)
return 1;
return 0;
}
/** ZSTD_compressSubBlock_multi() :
* Breaks super-block into multiple sub-blocks and compresses them.
* Entropy will be written to the first block.
* The following blocks will use repeat mode to compress.
* All sub-blocks are compressed blocks (no raw or rle blocks).
* @return : compressed size of the super block (which is multiple ZSTD blocks)
* Or 0 if it failed to compress. */
static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr,
const ZSTD_compressedBlockState_t* prevCBlock,
ZSTD_compressedBlockState_t* nextCBlock,
const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
const ZSTD_CCtx_params* cctxParams,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const int bmi2, U32 lastBlock,
void* workspace, size_t wkspSize)
{
const seqDef* const sstart = seqStorePtr->sequencesStart;
const seqDef* const send = seqStorePtr->sequences;
const seqDef* sp = sstart;
const BYTE* const lstart = seqStorePtr->litStart;
const BYTE* const lend = seqStorePtr->lit;
const BYTE* lp = lstart;
BYTE const* ip = (BYTE const*)src;
BYTE const* const iend = ip + srcSize;
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstCapacity;
BYTE* op = ostart;
const BYTE* llCodePtr = seqStorePtr->llCode;
const BYTE* mlCodePtr = seqStorePtr->mlCode;
const BYTE* ofCodePtr = seqStorePtr->ofCode;
size_t targetCBlockSize = cctxParams->targetCBlockSize;
size_t litSize, seqCount;
int writeLitEntropy = entropyMetadata->hufMetadata.hType == set_compressed;
int writeSeqEntropy = 1;
int lastSequence = 0;
DEBUGLOG(5, "ZSTD_compressSubBlock_multi (litSize=%u, nbSeq=%u)",
(unsigned)(lend-lp), (unsigned)(send-sstart));
litSize = 0;
seqCount = 0;
do {
size_t cBlockSizeEstimate = 0;
if (sstart == send) {
lastSequence = 1;
} else {
const seqDef* const sequence = sp + seqCount;
lastSequence = sequence == send - 1;
litSize += ZSTD_getSequenceLength(seqStorePtr, sequence).litLength;
seqCount++;
}
if (lastSequence) {
assert(lp <= lend);
assert(litSize <= (size_t)(lend - lp));
litSize = (size_t)(lend - lp);
}
/* I think there is an optimization opportunity here.
* Calling ZSTD_estimateSubBlockSize for every sequence can be wasteful
* since it recalculates estimate from scratch.
* For example, it would recount literal distribution and symbol codes every time.
*/
cBlockSizeEstimate = ZSTD_estimateSubBlockSize(lp, litSize, ofCodePtr, llCodePtr, mlCodePtr, seqCount,
&nextCBlock->entropy, entropyMetadata,
workspace, wkspSize, writeLitEntropy, writeSeqEntropy);
if (cBlockSizeEstimate > targetCBlockSize || lastSequence) {
int litEntropyWritten = 0;
int seqEntropyWritten = 0;
const size_t decompressedSize = ZSTD_seqDecompressedSize(seqStorePtr, sp, seqCount, litSize, lastSequence);
const size_t cSize = ZSTD_compressSubBlock(&nextCBlock->entropy, entropyMetadata,
sp, seqCount,
lp, litSize,
llCodePtr, mlCodePtr, ofCodePtr,
cctxParams,
op, oend-op,
bmi2, writeLitEntropy, writeSeqEntropy,
&litEntropyWritten, &seqEntropyWritten,
lastBlock && lastSequence);
FORWARD_IF_ERROR(cSize, "ZSTD_compressSubBlock failed");
if (cSize > 0 && cSize < decompressedSize) {
DEBUGLOG(5, "Committed the sub-block");
assert(ip + decompressedSize <= iend);
ip += decompressedSize;
sp += seqCount;
lp += litSize;
op += cSize;
llCodePtr += seqCount;
mlCodePtr += seqCount;
ofCodePtr += seqCount;
litSize = 0;
seqCount = 0;
/* Entropy only needs to be written once */
if (litEntropyWritten) {
writeLitEntropy = 0;
}
if (seqEntropyWritten) {
writeSeqEntropy = 0;
}
}
}
} while (!lastSequence);
if (writeLitEntropy) {
DEBUGLOG(5, "ZSTD_compressSubBlock_multi has literal entropy tables unwritten");
ZSTD_memcpy(&nextCBlock->entropy.huf, &prevCBlock->entropy.huf, sizeof(prevCBlock->entropy.huf));
}
if (writeSeqEntropy && ZSTD_needSequenceEntropyTables(&entropyMetadata->fseMetadata)) {
/* If we haven't written our entropy tables, then we've violated our contract and
* must emit an uncompressed block.
*/
DEBUGLOG(5, "ZSTD_compressSubBlock_multi has sequence entropy tables unwritten");
return 0;
}
if (ip < iend) {
size_t const cSize = ZSTD_noCompressBlock(op, oend - op, ip, iend - ip, lastBlock);
DEBUGLOG(5, "ZSTD_compressSubBlock_multi last sub-block uncompressed, %zu bytes", (size_t)(iend - ip));
FORWARD_IF_ERROR(cSize, "ZSTD_noCompressBlock failed");
assert(cSize != 0);
op += cSize;
/* We have to regenerate the repcodes because we've skipped some sequences */
if (sp < send) {
seqDef const* seq;
repcodes_t rep;
ZSTD_memcpy(&rep, prevCBlock->rep, sizeof(rep));
for (seq = sstart; seq < sp; ++seq) {
ZSTD_updateRep(rep.rep, seq->offBase, ZSTD_getSequenceLength(seqStorePtr, seq).litLength == 0);
}
ZSTD_memcpy(nextCBlock->rep, &rep, sizeof(rep));
}
}
DEBUGLOG(5, "ZSTD_compressSubBlock_multi compressed");
return op-ostart;
}
size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc,
void* dst, size_t dstCapacity,
void const* src, size_t srcSize,
unsigned lastBlock) {
ZSTD_entropyCTablesMetadata_t entropyMetadata;
FORWARD_IF_ERROR(ZSTD_buildBlockEntropyStats(&zc->seqStore,
&zc->blockState.prevCBlock->entropy,
&zc->blockState.nextCBlock->entropy,
&zc->appliedParams,
&entropyMetadata,
zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */), "");
return ZSTD_compressSubBlock_multi(&zc->seqStore,
zc->blockState.prevCBlock,
zc->blockState.nextCBlock,
&entropyMetadata,
&zc->appliedParams,
dst, dstCapacity,
src, srcSize,
zc->bmi2, lastBlock,
zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */);
}

32
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_compress_superblock.h vendored Normal file
View file

@ -0,0 +1,32 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_COMPRESS_ADVANCED_H
#define ZSTD_COMPRESS_ADVANCED_H
/*-*************************************
* Dependencies
***************************************/
#include "../zstd.h" /* ZSTD_CCtx */
/*-*************************************
* Target Compressed Block Size
***************************************/
/* ZSTD_compressSuperBlock() :
* Used to compress a super block when targetCBlockSize is being used.
* The given block will be compressed into multiple sub blocks that are around targetCBlockSize. */
size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc,
void* dst, size_t dstCapacity,
void const* src, size_t srcSize,
unsigned lastBlock);
#endif /* ZSTD_COMPRESS_ADVANCED_H */

742
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_cwksp.h vendored Normal file
View file

@ -0,0 +1,742 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_CWKSP_H
#define ZSTD_CWKSP_H
/*-*************************************
* Dependencies
***************************************/
#include "../common/allocations.h" /* ZSTD_customMalloc, ZSTD_customFree */
#include "../common/zstd_internal.h"
#include "../common/portability_macros.h"
#if defined (__cplusplus)
extern "C" {
#endif
/*-*************************************
* Constants
***************************************/
/* Since the workspace is effectively its own little malloc implementation /
* arena, when we run under ASAN, we should similarly insert redzones between
* each internal element of the workspace, so ASAN will catch overruns that
* reach outside an object but that stay inside the workspace.
*
* This defines the size of that redzone.
*/
#ifndef ZSTD_CWKSP_ASAN_REDZONE_SIZE
#define ZSTD_CWKSP_ASAN_REDZONE_SIZE 128
#endif
/* Set our tables and aligneds to align by 64 bytes */
#define ZSTD_CWKSP_ALIGNMENT_BYTES 64
/*-*************************************
* Structures
***************************************/
typedef enum {
ZSTD_cwksp_alloc_objects,
ZSTD_cwksp_alloc_aligned_init_once,
ZSTD_cwksp_alloc_aligned,
ZSTD_cwksp_alloc_buffers
} ZSTD_cwksp_alloc_phase_e;
/**
* Used to describe whether the workspace is statically allocated (and will not
* necessarily ever be freed), or if it's dynamically allocated and we can
* expect a well-formed caller to free this.
*/
typedef enum {
ZSTD_cwksp_dynamic_alloc,
ZSTD_cwksp_static_alloc
} ZSTD_cwksp_static_alloc_e;
/**
* Zstd fits all its internal datastructures into a single continuous buffer,
* so that it only needs to perform a single OS allocation (or so that a buffer
* can be provided to it and it can perform no allocations at all). This buffer
* is called the workspace.
*
* Several optimizations complicate that process of allocating memory ranges
* from this workspace for each internal datastructure:
*
* - These different internal datastructures have different setup requirements:
*
* - The static objects need to be cleared once and can then be trivially
* reused for each compression.
*
* - Various buffers don't need to be initialized at all--they are always
* written into before they're read.
*
* - The matchstate tables have a unique requirement that they don't need
* their memory to be totally cleared, but they do need the memory to have
* some bound, i.e., a guarantee that all values in the memory they've been
* allocated is less than some maximum value (which is the starting value
* for the indices that they will then use for compression). When this
* guarantee is provided to them, they can use the memory without any setup
* work. When it can't, they have to clear the area.
*
* - These buffers also have different alignment requirements.
*
* - We would like to reuse the objects in the workspace for multiple
* compressions without having to perform any expensive reallocation or
* reinitialization work.
*
* - We would like to be able to efficiently reuse the workspace across
* multiple compressions **even when the compression parameters change** and
* we need to resize some of the objects (where possible).
*
* To attempt to manage this buffer, given these constraints, the ZSTD_cwksp
* abstraction was created. It works as follows:
*
* Workspace Layout:
*
* [ ... workspace ... ]
* [objects][tables ->] free space [<- buffers][<- aligned][<- init once]
*
* The various objects that live in the workspace are divided into the
* following categories, and are allocated separately:
*
* - Static objects: this is optionally the enclosing ZSTD_CCtx or ZSTD_CDict,
* so that literally everything fits in a single buffer. Note: if present,
* this must be the first object in the workspace, since ZSTD_customFree{CCtx,
* CDict}() rely on a pointer comparison to see whether one or two frees are
* required.
*
* - Fixed size objects: these are fixed-size, fixed-count objects that are
* nonetheless "dynamically" allocated in the workspace so that we can
* control how they're initialized separately from the broader ZSTD_CCtx.
* Examples:
* - Entropy Workspace
* - 2 x ZSTD_compressedBlockState_t
* - CDict dictionary contents
*
* - Tables: these are any of several different datastructures (hash tables,
* chain tables, binary trees) that all respect a common format: they are
* uint32_t arrays, all of whose values are between 0 and (nextSrc - base).
* Their sizes depend on the cparams. These tables are 64-byte aligned.
*
* - Init once: these buffers require to be initialized at least once before
* use. They should be used when we want to skip memory initialization
* while not triggering memory checkers (like Valgrind) when reading from
* from this memory without writing to it first.
* These buffers should be used carefully as they might contain data
* from previous compressions.
* Buffers are aligned to 64 bytes.
*
* - Aligned: these buffers don't require any initialization before they're
* used. The user of the buffer should make sure they write into a buffer
* location before reading from it.
* Buffers are aligned to 64 bytes.
*
* - Buffers: these buffers are used for various purposes that don't require
* any alignment or initialization before they're used. This means they can
* be moved around at no cost for a new compression.
*
* Allocating Memory:
*
* The various types of objects must be allocated in order, so they can be
* correctly packed into the workspace buffer. That order is:
*
* 1. Objects
* 2. Init once / Tables
* 3. Aligned / Tables
* 4. Buffers / Tables
*
* Attempts to reserve objects of different types out of order will fail.
*/
typedef struct {
void* workspace;
void* workspaceEnd;
void* objectEnd;
void* tableEnd;
void* tableValidEnd;
void* allocStart;
void* initOnceStart;
BYTE allocFailed;
int workspaceOversizedDuration;
ZSTD_cwksp_alloc_phase_e phase;
ZSTD_cwksp_static_alloc_e isStatic;
} ZSTD_cwksp;
/*-*************************************
* Functions
***************************************/
MEM_STATIC size_t ZSTD_cwksp_available_space(ZSTD_cwksp* ws);
MEM_STATIC void* ZSTD_cwksp_initialAllocStart(ZSTD_cwksp* ws);
MEM_STATIC void ZSTD_cwksp_assert_internal_consistency(ZSTD_cwksp* ws) {
(void)ws;
assert(ws->workspace <= ws->objectEnd);
assert(ws->objectEnd <= ws->tableEnd);
assert(ws->objectEnd <= ws->tableValidEnd);
assert(ws->tableEnd <= ws->allocStart);
assert(ws->tableValidEnd <= ws->allocStart);
assert(ws->allocStart <= ws->workspaceEnd);
assert(ws->initOnceStart <= ZSTD_cwksp_initialAllocStart(ws));
assert(ws->workspace <= ws->initOnceStart);
#if ZSTD_MEMORY_SANITIZER
{
intptr_t const offset = __msan_test_shadow(ws->initOnceStart,
(U8*)ZSTD_cwksp_initialAllocStart(ws) - (U8*)ws->initOnceStart);
#if defined(ZSTD_MSAN_PRINT)
if(offset!=-1) {
__msan_print_shadow((U8*)ws->initOnceStart + offset - 8, 32);
}
#endif
assert(offset==-1);
};
#endif
}
/**
* Align must be a power of 2.
*/
MEM_STATIC size_t ZSTD_cwksp_align(size_t size, size_t const align) {
size_t const mask = align - 1;
assert((align & mask) == 0);
return (size + mask) & ~mask;
}
/**
* Use this to determine how much space in the workspace we will consume to
* allocate this object. (Normally it should be exactly the size of the object,
* but under special conditions, like ASAN, where we pad each object, it might
* be larger.)
*
* Since tables aren't currently redzoned, you don't need to call through this
* to figure out how much space you need for the matchState tables. Everything
* else is though.
*
* Do not use for sizing aligned buffers. Instead, use ZSTD_cwksp_aligned_alloc_size().
*/
MEM_STATIC size_t ZSTD_cwksp_alloc_size(size_t size) {
if (size == 0)
return 0;
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
return size + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
#else
return size;
#endif
}
/**
* Returns an adjusted alloc size that is the nearest larger multiple of 64 bytes.
* Used to determine the number of bytes required for a given "aligned".
*/
MEM_STATIC size_t ZSTD_cwksp_aligned_alloc_size(size_t size) {
return ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(size, ZSTD_CWKSP_ALIGNMENT_BYTES));
}
/**
* Returns the amount of additional space the cwksp must allocate
* for internal purposes (currently only alignment).
*/
MEM_STATIC size_t ZSTD_cwksp_slack_space_required(void) {
/* For alignment, the wksp will always allocate an additional 2*ZSTD_CWKSP_ALIGNMENT_BYTES
* bytes to align the beginning of tables section and end of buffers;
*/
size_t const slackSpace = ZSTD_CWKSP_ALIGNMENT_BYTES * 2;
return slackSpace;
}
/**
* Return the number of additional bytes required to align a pointer to the given number of bytes.
* alignBytes must be a power of two.
*/
MEM_STATIC size_t ZSTD_cwksp_bytes_to_align_ptr(void* ptr, const size_t alignBytes) {
size_t const alignBytesMask = alignBytes - 1;
size_t const bytes = (alignBytes - ((size_t)ptr & (alignBytesMask))) & alignBytesMask;
assert((alignBytes & alignBytesMask) == 0);
assert(bytes < alignBytes);
return bytes;
}
/**
* Returns the initial value for allocStart which is used to determine the position from
* which we can allocate from the end of the workspace.
*/
MEM_STATIC void* ZSTD_cwksp_initialAllocStart(ZSTD_cwksp* ws) {
return (void*)((size_t)ws->workspaceEnd & ~(ZSTD_CWKSP_ALIGNMENT_BYTES-1));
}
/**
* Internal function. Do not use directly.
* Reserves the given number of bytes within the aligned/buffer segment of the wksp,
* which counts from the end of the wksp (as opposed to the object/table segment).
*
* Returns a pointer to the beginning of that space.
*/
MEM_STATIC void*
ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t const bytes)
{
void* const alloc = (BYTE*)ws->allocStart - bytes;
void* const bottom = ws->tableEnd;
DEBUGLOG(5, "cwksp: reserving %p %zd bytes, %zd bytes remaining",
alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes);
ZSTD_cwksp_assert_internal_consistency(ws);
assert(alloc >= bottom);
if (alloc < bottom) {
DEBUGLOG(4, "cwksp: alloc failed!");
ws->allocFailed = 1;
return NULL;
}
/* the area is reserved from the end of wksp.
* If it overlaps with tableValidEnd, it voids guarantees on values' range */
if (alloc < ws->tableValidEnd) {
ws->tableValidEnd = alloc;
}
ws->allocStart = alloc;
return alloc;
}
/**
* Moves the cwksp to the next phase, and does any necessary allocations.
* cwksp initialization must necessarily go through each phase in order.
* Returns a 0 on success, or zstd error
*/
MEM_STATIC size_t
ZSTD_cwksp_internal_advance_phase(ZSTD_cwksp* ws, ZSTD_cwksp_alloc_phase_e phase)
{
assert(phase >= ws->phase);
if (phase > ws->phase) {
/* Going from allocating objects to allocating initOnce / tables */
if (ws->phase < ZSTD_cwksp_alloc_aligned_init_once &&
phase >= ZSTD_cwksp_alloc_aligned_init_once) {
ws->tableValidEnd = ws->objectEnd;
ws->initOnceStart = ZSTD_cwksp_initialAllocStart(ws);
{ /* Align the start of the tables to 64 bytes. Use [0, 63] bytes */
void *const alloc = ws->objectEnd;
size_t const bytesToAlign = ZSTD_cwksp_bytes_to_align_ptr(alloc, ZSTD_CWKSP_ALIGNMENT_BYTES);
void *const objectEnd = (BYTE *) alloc + bytesToAlign;
DEBUGLOG(5, "reserving table alignment addtl space: %zu", bytesToAlign);
RETURN_ERROR_IF(objectEnd > ws->workspaceEnd, memory_allocation,
"table phase - alignment initial allocation failed!");
ws->objectEnd = objectEnd;
ws->tableEnd = objectEnd; /* table area starts being empty */
if (ws->tableValidEnd < ws->tableEnd) {
ws->tableValidEnd = ws->tableEnd;
}
}
}
ws->phase = phase;
ZSTD_cwksp_assert_internal_consistency(ws);
}
return 0;
}
/**
* Returns whether this object/buffer/etc was allocated in this workspace.
*/
MEM_STATIC int ZSTD_cwksp_owns_buffer(const ZSTD_cwksp* ws, const void* ptr)
{
return (ptr != NULL) && (ws->workspace <= ptr) && (ptr < ws->workspaceEnd);
}
/**
* Internal function. Do not use directly.
*/
MEM_STATIC void*
ZSTD_cwksp_reserve_internal(ZSTD_cwksp* ws, size_t bytes, ZSTD_cwksp_alloc_phase_e phase)
{
void* alloc;
if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase)) || bytes == 0) {
return NULL;
}
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* over-reserve space */
bytes += 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
#endif
alloc = ZSTD_cwksp_reserve_internal_buffer_space(ws, bytes);
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on
* either size. */
if (alloc) {
alloc = (BYTE *)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
/* We need to keep the redzone poisoned while unpoisoning the bytes that
* are actually allocated. */
__asan_unpoison_memory_region(alloc, bytes - 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE);
}
}
#endif
return alloc;
}
/**
* Reserves and returns unaligned memory.
*/
MEM_STATIC BYTE* ZSTD_cwksp_reserve_buffer(ZSTD_cwksp* ws, size_t bytes)
{
return (BYTE*)ZSTD_cwksp_reserve_internal(ws, bytes, ZSTD_cwksp_alloc_buffers);
}
/**
* Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes).
* This memory has been initialized at least once in the past.
* This doesn't mean it has been initialized this time, and it might contain data from previous
* operations.
* The main usage is for algorithms that might need read access into uninitialized memory.
* The algorithm must maintain safety under these conditions and must make sure it doesn't
* leak any of the past data (directly or in side channels).
*/
MEM_STATIC void* ZSTD_cwksp_reserve_aligned_init_once(ZSTD_cwksp* ws, size_t bytes)
{
size_t const alignedBytes = ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES);
void* ptr = ZSTD_cwksp_reserve_internal(ws, alignedBytes, ZSTD_cwksp_alloc_aligned_init_once);
assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
if(ptr && ptr < ws->initOnceStart) {
/* We assume the memory following the current allocation is either:
* 1. Not usable as initOnce memory (end of workspace)
* 2. Another initOnce buffer that has been allocated before (and so was previously memset)
* 3. An ASAN redzone, in which case we don't want to write on it
* For these reasons it should be fine to not explicitly zero every byte up to ws->initOnceStart.
* Note that we assume here that MSAN and ASAN cannot run in the same time. */
ZSTD_memset(ptr, 0, MIN((size_t)((U8*)ws->initOnceStart - (U8*)ptr), alignedBytes));
ws->initOnceStart = ptr;
}
#if ZSTD_MEMORY_SANITIZER
assert(__msan_test_shadow(ptr, bytes) == -1);
#endif
return ptr;
}
/**
* Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes).
*/
MEM_STATIC void* ZSTD_cwksp_reserve_aligned(ZSTD_cwksp* ws, size_t bytes)
{
void* ptr = ZSTD_cwksp_reserve_internal(ws, ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES),
ZSTD_cwksp_alloc_aligned);
assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
return ptr;
}
/**
* Aligned on 64 bytes. These buffers have the special property that
* their values remain constrained, allowing us to re-use them without
* memset()-ing them.
*/
MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes)
{
const ZSTD_cwksp_alloc_phase_e phase = ZSTD_cwksp_alloc_aligned_init_once;
void* alloc;
void* end;
void* top;
/* We can only start allocating tables after we are done reserving space for objects at the
* start of the workspace */
if(ws->phase < phase) {
if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase))) {
return NULL;
}
}
alloc = ws->tableEnd;
end = (BYTE *)alloc + bytes;
top = ws->allocStart;
DEBUGLOG(5, "cwksp: reserving %p table %zd bytes, %zd bytes remaining",
alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes);
assert((bytes & (sizeof(U32)-1)) == 0);
ZSTD_cwksp_assert_internal_consistency(ws);
assert(end <= top);
if (end > top) {
DEBUGLOG(4, "cwksp: table alloc failed!");
ws->allocFailed = 1;
return NULL;
}
ws->tableEnd = end;
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
__asan_unpoison_memory_region(alloc, bytes);
}
#endif
assert((bytes & (ZSTD_CWKSP_ALIGNMENT_BYTES-1)) == 0);
assert(((size_t)alloc & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
return alloc;
}
/**
* Aligned on sizeof(void*).
* Note : should happen only once, at workspace first initialization
*/
MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes)
{
size_t const roundedBytes = ZSTD_cwksp_align(bytes, sizeof(void*));
void* alloc = ws->objectEnd;
void* end = (BYTE*)alloc + roundedBytes;
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* over-reserve space */
end = (BYTE *)end + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
#endif
DEBUGLOG(4,
"cwksp: reserving %p object %zd bytes (rounded to %zd), %zd bytes remaining",
alloc, bytes, roundedBytes, ZSTD_cwksp_available_space(ws) - roundedBytes);
assert((size_t)alloc % ZSTD_ALIGNOF(void*) == 0);
assert(bytes % ZSTD_ALIGNOF(void*) == 0);
ZSTD_cwksp_assert_internal_consistency(ws);
/* we must be in the first phase, no advance is possible */
if (ws->phase != ZSTD_cwksp_alloc_objects || end > ws->workspaceEnd) {
DEBUGLOG(3, "cwksp: object alloc failed!");
ws->allocFailed = 1;
return NULL;
}
ws->objectEnd = end;
ws->tableEnd = end;
ws->tableValidEnd = end;
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on
* either size. */
alloc = (BYTE*)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
__asan_unpoison_memory_region(alloc, bytes);
}
#endif
return alloc;
}
MEM_STATIC void ZSTD_cwksp_mark_tables_dirty(ZSTD_cwksp* ws)
{
DEBUGLOG(4, "cwksp: ZSTD_cwksp_mark_tables_dirty");
#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
/* To validate that the table re-use logic is sound, and that we don't
* access table space that we haven't cleaned, we re-"poison" the table
* space every time we mark it dirty.
* Since tableValidEnd space and initOnce space may overlap we don't poison
* the initOnce portion as it break its promise. This means that this poisoning
* check isn't always applied fully. */
{
size_t size = (BYTE*)ws->tableValidEnd - (BYTE*)ws->objectEnd;
assert(__msan_test_shadow(ws->objectEnd, size) == -1);
if((BYTE*)ws->tableValidEnd < (BYTE*)ws->initOnceStart) {
__msan_poison(ws->objectEnd, size);
} else {
assert(ws->initOnceStart >= ws->objectEnd);
__msan_poison(ws->objectEnd, (BYTE*)ws->initOnceStart - (BYTE*)ws->objectEnd);
}
}
#endif
assert(ws->tableValidEnd >= ws->objectEnd);
assert(ws->tableValidEnd <= ws->allocStart);
ws->tableValidEnd = ws->objectEnd;
ZSTD_cwksp_assert_internal_consistency(ws);
}
MEM_STATIC void ZSTD_cwksp_mark_tables_clean(ZSTD_cwksp* ws) {
DEBUGLOG(4, "cwksp: ZSTD_cwksp_mark_tables_clean");
assert(ws->tableValidEnd >= ws->objectEnd);
assert(ws->tableValidEnd <= ws->allocStart);
if (ws->tableValidEnd < ws->tableEnd) {
ws->tableValidEnd = ws->tableEnd;
}
ZSTD_cwksp_assert_internal_consistency(ws);
}
/**
* Zero the part of the allocated tables not already marked clean.
*/
MEM_STATIC void ZSTD_cwksp_clean_tables(ZSTD_cwksp* ws) {
DEBUGLOG(4, "cwksp: ZSTD_cwksp_clean_tables");
assert(ws->tableValidEnd >= ws->objectEnd);
assert(ws->tableValidEnd <= ws->allocStart);
if (ws->tableValidEnd < ws->tableEnd) {
ZSTD_memset(ws->tableValidEnd, 0, (size_t)((BYTE*)ws->tableEnd - (BYTE*)ws->tableValidEnd));
}
ZSTD_cwksp_mark_tables_clean(ws);
}
/**
* Invalidates table allocations.
* All other allocations remain valid.
*/
MEM_STATIC void ZSTD_cwksp_clear_tables(ZSTD_cwksp* ws) {
DEBUGLOG(4, "cwksp: clearing tables!");
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* We don't do this when the workspace is statically allocated, because
* when that is the case, we have no capability to hook into the end of the
* workspace's lifecycle to unpoison the memory.
*/
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
size_t size = (BYTE*)ws->tableValidEnd - (BYTE*)ws->objectEnd;
__asan_poison_memory_region(ws->objectEnd, size);
}
#endif
ws->tableEnd = ws->objectEnd;
ZSTD_cwksp_assert_internal_consistency(ws);
}
/**
* Invalidates all buffer, aligned, and table allocations.
* Object allocations remain valid.
*/
MEM_STATIC void ZSTD_cwksp_clear(ZSTD_cwksp* ws) {
DEBUGLOG(4, "cwksp: clearing!");
#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
/* To validate that the context re-use logic is sound, and that we don't
* access stuff that this compression hasn't initialized, we re-"poison"
* the workspace except for the areas in which we expect memory re-use
* without initialization (objects, valid tables area and init once
* memory). */
{
if((BYTE*)ws->tableValidEnd < (BYTE*)ws->initOnceStart) {
size_t size = (BYTE*)ws->initOnceStart - (BYTE*)ws->tableValidEnd;
__msan_poison(ws->tableValidEnd, size);
}
}
#endif
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* We don't do this when the workspace is statically allocated, because
* when that is the case, we have no capability to hook into the end of the
* workspace's lifecycle to unpoison the memory.
*/
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
size_t size = (BYTE*)ws->workspaceEnd - (BYTE*)ws->objectEnd;
__asan_poison_memory_region(ws->objectEnd, size);
}
#endif
ws->tableEnd = ws->objectEnd;
ws->allocStart = ZSTD_cwksp_initialAllocStart(ws);
ws->allocFailed = 0;
if (ws->phase > ZSTD_cwksp_alloc_aligned_init_once) {
ws->phase = ZSTD_cwksp_alloc_aligned_init_once;
}
ZSTD_cwksp_assert_internal_consistency(ws);
}
/**
* The provided workspace takes ownership of the buffer [start, start+size).
* Any existing values in the workspace are ignored (the previously managed
* buffer, if present, must be separately freed).
*/
MEM_STATIC void ZSTD_cwksp_init(ZSTD_cwksp* ws, void* start, size_t size, ZSTD_cwksp_static_alloc_e isStatic) {
DEBUGLOG(4, "cwksp: init'ing workspace with %zd bytes", size);
assert(((size_t)start & (sizeof(void*)-1)) == 0); /* ensure correct alignment */
ws->workspace = start;
ws->workspaceEnd = (BYTE*)start + size;
ws->objectEnd = ws->workspace;
ws->tableValidEnd = ws->objectEnd;
ws->initOnceStart = ZSTD_cwksp_initialAllocStart(ws);
ws->phase = ZSTD_cwksp_alloc_objects;
ws->isStatic = isStatic;
ZSTD_cwksp_clear(ws);
ws->workspaceOversizedDuration = 0;
ZSTD_cwksp_assert_internal_consistency(ws);
}
MEM_STATIC size_t ZSTD_cwksp_create(ZSTD_cwksp* ws, size_t size, ZSTD_customMem customMem) {
void* workspace = ZSTD_customMalloc(size, customMem);
DEBUGLOG(4, "cwksp: creating new workspace with %zd bytes", size);
RETURN_ERROR_IF(workspace == NULL, memory_allocation, "NULL pointer!");
ZSTD_cwksp_init(ws, workspace, size, ZSTD_cwksp_dynamic_alloc);
return 0;
}
MEM_STATIC void ZSTD_cwksp_free(ZSTD_cwksp* ws, ZSTD_customMem customMem) {
void *ptr = ws->workspace;
DEBUGLOG(4, "cwksp: freeing workspace");
ZSTD_memset(ws, 0, sizeof(ZSTD_cwksp));
ZSTD_customFree(ptr, customMem);
}
/**
* Moves the management of a workspace from one cwksp to another. The src cwksp
* is left in an invalid state (src must be re-init()'ed before it's used again).
*/
MEM_STATIC void ZSTD_cwksp_move(ZSTD_cwksp* dst, ZSTD_cwksp* src) {
*dst = *src;
ZSTD_memset(src, 0, sizeof(ZSTD_cwksp));
}
MEM_STATIC size_t ZSTD_cwksp_sizeof(const ZSTD_cwksp* ws) {
return (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->workspace);
}
MEM_STATIC size_t ZSTD_cwksp_used(const ZSTD_cwksp* ws) {
return (size_t)((BYTE*)ws->tableEnd - (BYTE*)ws->workspace)
+ (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->allocStart);
}
MEM_STATIC int ZSTD_cwksp_reserve_failed(const ZSTD_cwksp* ws) {
return ws->allocFailed;
}
/*-*************************************
* Functions Checking Free Space
***************************************/
/* ZSTD_alignmentSpaceWithinBounds() :
* Returns if the estimated space needed for a wksp is within an acceptable limit of the
* actual amount of space used.
*/
MEM_STATIC int ZSTD_cwksp_estimated_space_within_bounds(const ZSTD_cwksp *const ws, size_t const estimatedSpace) {
/* We have an alignment space between objects and tables between tables and buffers, so we can have up to twice
* the alignment bytes difference between estimation and actual usage */
return (estimatedSpace - ZSTD_cwksp_slack_space_required()) <= ZSTD_cwksp_used(ws) &&
ZSTD_cwksp_used(ws) <= estimatedSpace;
}
MEM_STATIC size_t ZSTD_cwksp_available_space(ZSTD_cwksp* ws) {
return (size_t)((BYTE*)ws->allocStart - (BYTE*)ws->tableEnd);
}
MEM_STATIC int ZSTD_cwksp_check_available(ZSTD_cwksp* ws, size_t additionalNeededSpace) {
return ZSTD_cwksp_available_space(ws) >= additionalNeededSpace;
}
MEM_STATIC int ZSTD_cwksp_check_too_large(ZSTD_cwksp* ws, size_t additionalNeededSpace) {
return ZSTD_cwksp_check_available(
ws, additionalNeededSpace * ZSTD_WORKSPACETOOLARGE_FACTOR);
}
MEM_STATIC int ZSTD_cwksp_check_wasteful(ZSTD_cwksp* ws, size_t additionalNeededSpace) {
return ZSTD_cwksp_check_too_large(ws, additionalNeededSpace)
&& ws->workspaceOversizedDuration > ZSTD_WORKSPACETOOLARGE_MAXDURATION;
}
MEM_STATIC void ZSTD_cwksp_bump_oversized_duration(
ZSTD_cwksp* ws, size_t additionalNeededSpace) {
if (ZSTD_cwksp_check_too_large(ws, additionalNeededSpace)) {
ws->workspaceOversizedDuration++;
} else {
ws->workspaceOversizedDuration = 0;
}
}
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_CWKSP_H */

758
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_double_fast.c vendored Normal file
View file

@ -0,0 +1,758 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#include "zstd_compress_internal.h"
#include "zstd_double_fast.h"
static void ZSTD_fillDoubleHashTableForCDict(ZSTD_matchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashLarge = ms->hashTable;
U32 const hBitsL = cParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
U32 const mls = cParams->minMatch;
U32* const hashSmall = ms->chainTable;
U32 const hBitsS = cParams->chainLog + ZSTD_SHORT_CACHE_TAG_BITS;
const BYTE* const base = ms->window.base;
const BYTE* ip = base + ms->nextToUpdate;
const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
const U32 fastHashFillStep = 3;
/* Always insert every fastHashFillStep position into the hash tables.
* Insert the other positions into the large hash table if their entry
* is empty.
*/
for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) {
U32 const curr = (U32)(ip - base);
U32 i;
for (i = 0; i < fastHashFillStep; ++i) {
size_t const smHashAndTag = ZSTD_hashPtr(ip + i, hBitsS, mls);
size_t const lgHashAndTag = ZSTD_hashPtr(ip + i, hBitsL, 8);
if (i == 0) {
ZSTD_writeTaggedIndex(hashSmall, smHashAndTag, curr + i);
}
if (i == 0 || hashLarge[lgHashAndTag >> ZSTD_SHORT_CACHE_TAG_BITS] == 0) {
ZSTD_writeTaggedIndex(hashLarge, lgHashAndTag, curr + i);
}
/* Only load extra positions for ZSTD_dtlm_full */
if (dtlm == ZSTD_dtlm_fast)
break;
} }
}
static void ZSTD_fillDoubleHashTableForCCtx(ZSTD_matchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashLarge = ms->hashTable;
U32 const hBitsL = cParams->hashLog;
U32 const mls = cParams->minMatch;
U32* const hashSmall = ms->chainTable;
U32 const hBitsS = cParams->chainLog;
const BYTE* const base = ms->window.base;
const BYTE* ip = base + ms->nextToUpdate;
const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
const U32 fastHashFillStep = 3;
/* Always insert every fastHashFillStep position into the hash tables.
* Insert the other positions into the large hash table if their entry
* is empty.
*/
for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) {
U32 const curr = (U32)(ip - base);
U32 i;
for (i = 0; i < fastHashFillStep; ++i) {
size_t const smHash = ZSTD_hashPtr(ip + i, hBitsS, mls);
size_t const lgHash = ZSTD_hashPtr(ip + i, hBitsL, 8);
if (i == 0)
hashSmall[smHash] = curr + i;
if (i == 0 || hashLarge[lgHash] == 0)
hashLarge[lgHash] = curr + i;
/* Only load extra positions for ZSTD_dtlm_full */
if (dtlm == ZSTD_dtlm_fast)
break;
} }
}
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp)
{
if (tfp == ZSTD_tfp_forCDict) {
ZSTD_fillDoubleHashTableForCDict(ms, end, dtlm);
} else {
ZSTD_fillDoubleHashTableForCCtx(ms, end, dtlm);
}
}
FORCE_INLINE_TEMPLATE
size_t ZSTD_compressBlock_doubleFast_noDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize, U32 const mls /* template */)
{
ZSTD_compressionParameters const* cParams = &ms->cParams;
U32* const hashLong = ms->hashTable;
const U32 hBitsL = cParams->hashLog;
U32* const hashSmall = ms->chainTable;
const U32 hBitsS = cParams->chainLog;
const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
const BYTE* anchor = istart;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
/* presumes that, if there is a dictionary, it must be using Attach mode */
const U32 prefixLowestIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
const BYTE* const prefixLowest = base + prefixLowestIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
U32 offsetSaved1 = 0, offsetSaved2 = 0;
size_t mLength;
U32 offset;
U32 curr;
/* how many positions to search before increasing step size */
const size_t kStepIncr = 1 << kSearchStrength;
/* the position at which to increment the step size if no match is found */
const BYTE* nextStep;
size_t step; /* the current step size */
size_t hl0; /* the long hash at ip */
size_t hl1; /* the long hash at ip1 */
U32 idxl0; /* the long match index for ip */
U32 idxl1; /* the long match index for ip1 */
const BYTE* matchl0; /* the long match for ip */
const BYTE* matchs0; /* the short match for ip */
const BYTE* matchl1; /* the long match for ip1 */
const BYTE* ip = istart; /* the current position */
const BYTE* ip1; /* the next position */
DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_noDict_generic");
/* init */
ip += ((ip - prefixLowest) == 0);
{
U32 const current = (U32)(ip - base);
U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, current, cParams->windowLog);
U32 const maxRep = current - windowLow;
if (offset_2 > maxRep) offsetSaved2 = offset_2, offset_2 = 0;
if (offset_1 > maxRep) offsetSaved1 = offset_1, offset_1 = 0;
}
/* Outer Loop: one iteration per match found and stored */
while (1) {
step = 1;
nextStep = ip + kStepIncr;
ip1 = ip + step;
if (ip1 > ilimit) {
goto _cleanup;
}
hl0 = ZSTD_hashPtr(ip, hBitsL, 8);
idxl0 = hashLong[hl0];
matchl0 = base + idxl0;
/* Inner Loop: one iteration per search / position */
do {
const size_t hs0 = ZSTD_hashPtr(ip, hBitsS, mls);
const U32 idxs0 = hashSmall[hs0];
curr = (U32)(ip-base);
matchs0 = base + idxs0;
hashLong[hl0] = hashSmall[hs0] = curr; /* update hash tables */
/* check noDict repcode */
if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) {
mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
ip++;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
goto _match_stored;
}
hl1 = ZSTD_hashPtr(ip1, hBitsL, 8);
if (idxl0 > prefixLowestIndex) {
/* check prefix long match */
if (MEM_read64(matchl0) == MEM_read64(ip)) {
mLength = ZSTD_count(ip+8, matchl0+8, iend) + 8;
offset = (U32)(ip-matchl0);
while (((ip>anchor) & (matchl0>prefixLowest)) && (ip[-1] == matchl0[-1])) { ip--; matchl0--; mLength++; } /* catch up */
goto _match_found;
}
}
idxl1 = hashLong[hl1];
matchl1 = base + idxl1;
if (idxs0 > prefixLowestIndex) {
/* check prefix short match */
if (MEM_read32(matchs0) == MEM_read32(ip)) {
goto _search_next_long;
}
}
if (ip1 >= nextStep) {
PREFETCH_L1(ip1 + 64);
PREFETCH_L1(ip1 + 128);
step++;
nextStep += kStepIncr;
}
ip = ip1;
ip1 += step;
hl0 = hl1;
idxl0 = idxl1;
matchl0 = matchl1;
#if defined(__aarch64__)
PREFETCH_L1(ip+256);
#endif
} while (ip1 <= ilimit);
_cleanup:
/* If offset_1 started invalid (offsetSaved1 != 0) and became valid (offset_1 != 0),
* rotate saved offsets. See comment in ZSTD_compressBlock_fast_noDict for more context. */
offsetSaved2 = ((offsetSaved1 != 0) && (offset_1 != 0)) ? offsetSaved1 : offsetSaved2;
/* save reps for next block */
rep[0] = offset_1 ? offset_1 : offsetSaved1;
rep[1] = offset_2 ? offset_2 : offsetSaved2;
/* Return the last literals size */
return (size_t)(iend - anchor);
_search_next_long:
/* check prefix long +1 match */
if (idxl1 > prefixLowestIndex) {
if (MEM_read64(matchl1) == MEM_read64(ip1)) {
ip = ip1;
mLength = ZSTD_count(ip+8, matchl1+8, iend) + 8;
offset = (U32)(ip-matchl1);
while (((ip>anchor) & (matchl1>prefixLowest)) && (ip[-1] == matchl1[-1])) { ip--; matchl1--; mLength++; } /* catch up */
goto _match_found;
}
}
/* if no long +1 match, explore the short match we found */
mLength = ZSTD_count(ip+4, matchs0+4, iend) + 4;
offset = (U32)(ip - matchs0);
while (((ip>anchor) & (matchs0>prefixLowest)) && (ip[-1] == matchs0[-1])) { ip--; matchs0--; mLength++; } /* catch up */
/* fall-through */
_match_found: /* requires ip, offset, mLength */
offset_2 = offset_1;
offset_1 = offset;
if (step < 4) {
/* It is unsafe to write this value back to the hashtable when ip1 is
* greater than or equal to the new ip we will have after we're done
* processing this match. Rather than perform that test directly
* (ip1 >= ip + mLength), which costs speed in practice, we do a simpler
* more predictable test. The minmatch even if we take a short match is
* 4 bytes, so as long as step, the distance between ip and ip1
* (initially) is less than 4, we know ip1 < new ip. */
hashLong[hl1] = (U32)(ip1 - base);
}
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
_match_stored:
/* match found */
ip += mLength;
anchor = ip;
if (ip <= ilimit) {
/* Complementary insertion */
/* done after iLimit test, as candidates could be > iend-8 */
{ U32 const indexToInsert = curr+2;
hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
}
/* check immediate repcode */
while ( (ip <= ilimit)
&& ( (offset_2>0)
& (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
/* store sequence */
size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; /* swap offset_2 <=> offset_1 */
hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base);
hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base);
ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, rLength);
ip += rLength;
anchor = ip;
continue; /* faster when present ... (?) */
}
}
}
}
FORCE_INLINE_TEMPLATE
size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
U32 const mls /* template */)
{
ZSTD_compressionParameters const* cParams = &ms->cParams;
U32* const hashLong = ms->hashTable;
const U32 hBitsL = cParams->hashLog;
U32* const hashSmall = ms->chainTable;
const U32 hBitsS = cParams->chainLog;
const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip = istart;
const BYTE* anchor = istart;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
/* presumes that, if there is a dictionary, it must be using Attach mode */
const U32 prefixLowestIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
const BYTE* const prefixLowest = base + prefixLowestIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
const ZSTD_matchState_t* const dms = ms->dictMatchState;
const ZSTD_compressionParameters* const dictCParams = &dms->cParams;
const U32* const dictHashLong = dms->hashTable;
const U32* const dictHashSmall = dms->chainTable;
const U32 dictStartIndex = dms->window.dictLimit;
const BYTE* const dictBase = dms->window.base;
const BYTE* const dictStart = dictBase + dictStartIndex;
const BYTE* const dictEnd = dms->window.nextSrc;
const U32 dictIndexDelta = prefixLowestIndex - (U32)(dictEnd - dictBase);
const U32 dictHBitsL = dictCParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
const U32 dictHBitsS = dictCParams->chainLog + ZSTD_SHORT_CACHE_TAG_BITS;
const U32 dictAndPrefixLength = (U32)((ip - prefixLowest) + (dictEnd - dictStart));
DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_dictMatchState_generic");
/* if a dictionary is attached, it must be within window range */
assert(ms->window.dictLimit + (1U << cParams->windowLog) >= endIndex);
if (ms->prefetchCDictTables) {
size_t const hashTableBytes = (((size_t)1) << dictCParams->hashLog) * sizeof(U32);
size_t const chainTableBytes = (((size_t)1) << dictCParams->chainLog) * sizeof(U32);
PREFETCH_AREA(dictHashLong, hashTableBytes)
PREFETCH_AREA(dictHashSmall, chainTableBytes)
}
/* init */
ip += (dictAndPrefixLength == 0);
/* dictMatchState repCode checks don't currently handle repCode == 0
* disabling. */
assert(offset_1 <= dictAndPrefixLength);
assert(offset_2 <= dictAndPrefixLength);
/* Main Search Loop */
while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */
size_t mLength;
U32 offset;
size_t const h2 = ZSTD_hashPtr(ip, hBitsL, 8);
size_t const h = ZSTD_hashPtr(ip, hBitsS, mls);
size_t const dictHashAndTagL = ZSTD_hashPtr(ip, dictHBitsL, 8);
size_t const dictHashAndTagS = ZSTD_hashPtr(ip, dictHBitsS, mls);
U32 const dictMatchIndexAndTagL = dictHashLong[dictHashAndTagL >> ZSTD_SHORT_CACHE_TAG_BITS];
U32 const dictMatchIndexAndTagS = dictHashSmall[dictHashAndTagS >> ZSTD_SHORT_CACHE_TAG_BITS];
int const dictTagsMatchL = ZSTD_comparePackedTags(dictMatchIndexAndTagL, dictHashAndTagL);
int const dictTagsMatchS = ZSTD_comparePackedTags(dictMatchIndexAndTagS, dictHashAndTagS);
U32 const curr = (U32)(ip-base);
U32 const matchIndexL = hashLong[h2];
U32 matchIndexS = hashSmall[h];
const BYTE* matchLong = base + matchIndexL;
const BYTE* match = base + matchIndexS;
const U32 repIndex = curr + 1 - offset_1;
const BYTE* repMatch = (repIndex < prefixLowestIndex) ?
dictBase + (repIndex - dictIndexDelta) :
base + repIndex;
hashLong[h2] = hashSmall[h] = curr; /* update hash tables */
/* check repcode */
if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */)
&& (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
ip++;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
goto _match_stored;
}
if (matchIndexL > prefixLowestIndex) {
/* check prefix long match */
if (MEM_read64(matchLong) == MEM_read64(ip)) {
mLength = ZSTD_count(ip+8, matchLong+8, iend) + 8;
offset = (U32)(ip-matchLong);
while (((ip>anchor) & (matchLong>prefixLowest)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
goto _match_found;
}
} else if (dictTagsMatchL) {
/* check dictMatchState long match */
U32 const dictMatchIndexL = dictMatchIndexAndTagL >> ZSTD_SHORT_CACHE_TAG_BITS;
const BYTE* dictMatchL = dictBase + dictMatchIndexL;
assert(dictMatchL < dictEnd);
if (dictMatchL > dictStart && MEM_read64(dictMatchL) == MEM_read64(ip)) {
mLength = ZSTD_count_2segments(ip+8, dictMatchL+8, iend, dictEnd, prefixLowest) + 8;
offset = (U32)(curr - dictMatchIndexL - dictIndexDelta);
while (((ip>anchor) & (dictMatchL>dictStart)) && (ip[-1] == dictMatchL[-1])) { ip--; dictMatchL--; mLength++; } /* catch up */
goto _match_found;
} }
if (matchIndexS > prefixLowestIndex) {
/* check prefix short match */
if (MEM_read32(match) == MEM_read32(ip)) {
goto _search_next_long;
}
} else if (dictTagsMatchS) {
/* check dictMatchState short match */
U32 const dictMatchIndexS = dictMatchIndexAndTagS >> ZSTD_SHORT_CACHE_TAG_BITS;
match = dictBase + dictMatchIndexS;
matchIndexS = dictMatchIndexS + dictIndexDelta;
if (match > dictStart && MEM_read32(match) == MEM_read32(ip)) {
goto _search_next_long;
} }
ip += ((ip-anchor) >> kSearchStrength) + 1;
#if defined(__aarch64__)
PREFETCH_L1(ip+256);
#endif
continue;
_search_next_long:
{ size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
size_t const dictHashAndTagL3 = ZSTD_hashPtr(ip+1, dictHBitsL, 8);
U32 const matchIndexL3 = hashLong[hl3];
U32 const dictMatchIndexAndTagL3 = dictHashLong[dictHashAndTagL3 >> ZSTD_SHORT_CACHE_TAG_BITS];
int const dictTagsMatchL3 = ZSTD_comparePackedTags(dictMatchIndexAndTagL3, dictHashAndTagL3);
const BYTE* matchL3 = base + matchIndexL3;
hashLong[hl3] = curr + 1;
/* check prefix long +1 match */
if (matchIndexL3 > prefixLowestIndex) {
if (MEM_read64(matchL3) == MEM_read64(ip+1)) {
mLength = ZSTD_count(ip+9, matchL3+8, iend) + 8;
ip++;
offset = (U32)(ip-matchL3);
while (((ip>anchor) & (matchL3>prefixLowest)) && (ip[-1] == matchL3[-1])) { ip--; matchL3--; mLength++; } /* catch up */
goto _match_found;
}
} else if (dictTagsMatchL3) {
/* check dict long +1 match */
U32 const dictMatchIndexL3 = dictMatchIndexAndTagL3 >> ZSTD_SHORT_CACHE_TAG_BITS;
const BYTE* dictMatchL3 = dictBase + dictMatchIndexL3;
assert(dictMatchL3 < dictEnd);
if (dictMatchL3 > dictStart && MEM_read64(dictMatchL3) == MEM_read64(ip+1)) {
mLength = ZSTD_count_2segments(ip+1+8, dictMatchL3+8, iend, dictEnd, prefixLowest) + 8;
ip++;
offset = (U32)(curr + 1 - dictMatchIndexL3 - dictIndexDelta);
while (((ip>anchor) & (dictMatchL3>dictStart)) && (ip[-1] == dictMatchL3[-1])) { ip--; dictMatchL3--; mLength++; } /* catch up */
goto _match_found;
} } }
/* if no long +1 match, explore the short match we found */
if (matchIndexS < prefixLowestIndex) {
mLength = ZSTD_count_2segments(ip+4, match+4, iend, dictEnd, prefixLowest) + 4;
offset = (U32)(curr - matchIndexS);
while (((ip>anchor) & (match>dictStart)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
} else {
mLength = ZSTD_count(ip+4, match+4, iend) + 4;
offset = (U32)(ip - match);
while (((ip>anchor) & (match>prefixLowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
}
_match_found:
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
_match_stored:
/* match found */
ip += mLength;
anchor = ip;
if (ip <= ilimit) {
/* Complementary insertion */
/* done after iLimit test, as candidates could be > iend-8 */
{ U32 const indexToInsert = curr+2;
hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
}
/* check immediate repcode */
while (ip <= ilimit) {
U32 const current2 = (U32)(ip-base);
U32 const repIndex2 = current2 - offset_2;
const BYTE* repMatch2 = repIndex2 < prefixLowestIndex ?
dictBase + repIndex2 - dictIndexDelta :
base + repIndex2;
if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */)
&& (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
const BYTE* const repEnd2 = repIndex2 < prefixLowestIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixLowest) + 4;
U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
ip += repLength2;
anchor = ip;
continue;
}
break;
}
}
} /* while (ip < ilimit) */
/* save reps for next block */
rep[0] = offset_1;
rep[1] = offset_2;
/* Return the last literals size */
return (size_t)(iend - anchor);
}
#define ZSTD_GEN_DFAST_FN(dictMode, mls) \
static size_t ZSTD_compressBlock_doubleFast_##dictMode##_##mls( \
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \
void const* src, size_t srcSize) \
{ \
return ZSTD_compressBlock_doubleFast_##dictMode##_generic(ms, seqStore, rep, src, srcSize, mls); \
}
ZSTD_GEN_DFAST_FN(noDict, 4)
ZSTD_GEN_DFAST_FN(noDict, 5)
ZSTD_GEN_DFAST_FN(noDict, 6)
ZSTD_GEN_DFAST_FN(noDict, 7)
ZSTD_GEN_DFAST_FN(dictMatchState, 4)
ZSTD_GEN_DFAST_FN(dictMatchState, 5)
ZSTD_GEN_DFAST_FN(dictMatchState, 6)
ZSTD_GEN_DFAST_FN(dictMatchState, 7)
size_t ZSTD_compressBlock_doubleFast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
const U32 mls = ms->cParams.minMatch;
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_doubleFast_noDict_4(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_doubleFast_noDict_5(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_doubleFast_noDict_6(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_doubleFast_noDict_7(ms, seqStore, rep, src, srcSize);
}
}
size_t ZSTD_compressBlock_doubleFast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
const U32 mls = ms->cParams.minMatch;
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_doubleFast_dictMatchState_4(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_doubleFast_dictMatchState_5(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_doubleFast_dictMatchState_6(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_doubleFast_dictMatchState_7(ms, seqStore, rep, src, srcSize);
}
}
static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
U32 const mls /* template */)
{
ZSTD_compressionParameters const* cParams = &ms->cParams;
U32* const hashLong = ms->hashTable;
U32 const hBitsL = cParams->hashLog;
U32* const hashSmall = ms->chainTable;
U32 const hBitsS = cParams->chainLog;
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip = istart;
const BYTE* anchor = istart;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - 8;
const BYTE* const base = ms->window.base;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
const U32 lowLimit = ZSTD_getLowestMatchIndex(ms, endIndex, cParams->windowLog);
const U32 dictStartIndex = lowLimit;
const U32 dictLimit = ms->window.dictLimit;
const U32 prefixStartIndex = (dictLimit > lowLimit) ? dictLimit : lowLimit;
const BYTE* const prefixStart = base + prefixStartIndex;
const BYTE* const dictBase = ms->window.dictBase;
const BYTE* const dictStart = dictBase + dictStartIndex;
const BYTE* const dictEnd = dictBase + prefixStartIndex;
U32 offset_1=rep[0], offset_2=rep[1];
DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_extDict_generic (srcSize=%zu)", srcSize);
/* if extDict is invalidated due to maxDistance, switch to "regular" variant */
if (prefixStartIndex == dictStartIndex)
return ZSTD_compressBlock_doubleFast(ms, seqStore, rep, src, srcSize);
/* Search Loop */
while (ip < ilimit) { /* < instead of <=, because (ip+1) */
const size_t hSmall = ZSTD_hashPtr(ip, hBitsS, mls);
const U32 matchIndex = hashSmall[hSmall];
const BYTE* const matchBase = matchIndex < prefixStartIndex ? dictBase : base;
const BYTE* match = matchBase + matchIndex;
const size_t hLong = ZSTD_hashPtr(ip, hBitsL, 8);
const U32 matchLongIndex = hashLong[hLong];
const BYTE* const matchLongBase = matchLongIndex < prefixStartIndex ? dictBase : base;
const BYTE* matchLong = matchLongBase + matchLongIndex;
const U32 curr = (U32)(ip-base);
const U32 repIndex = curr + 1 - offset_1; /* offset_1 expected <= curr +1 */
const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base;
const BYTE* const repMatch = repBase + repIndex;
size_t mLength;
hashSmall[hSmall] = hashLong[hLong] = curr; /* update hash table */
if ((((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex doesn't overlap dict + prefix */
& (offset_1 <= curr+1 - dictStartIndex)) /* note: we are searching at curr+1 */
&& (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
ip++;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
} else {
if ((matchLongIndex > dictStartIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
const BYTE* const matchEnd = matchLongIndex < prefixStartIndex ? dictEnd : iend;
const BYTE* const lowMatchPtr = matchLongIndex < prefixStartIndex ? dictStart : prefixStart;
U32 offset;
mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, prefixStart) + 8;
offset = curr - matchLongIndex;
while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
} else if ((matchIndex > dictStartIndex) && (MEM_read32(match) == MEM_read32(ip))) {
size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
U32 const matchIndex3 = hashLong[h3];
const BYTE* const match3Base = matchIndex3 < prefixStartIndex ? dictBase : base;
const BYTE* match3 = match3Base + matchIndex3;
U32 offset;
hashLong[h3] = curr + 1;
if ( (matchIndex3 > dictStartIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) {
const BYTE* const matchEnd = matchIndex3 < prefixStartIndex ? dictEnd : iend;
const BYTE* const lowMatchPtr = matchIndex3 < prefixStartIndex ? dictStart : prefixStart;
mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, prefixStart) + 8;
ip++;
offset = curr+1 - matchIndex3;
while (((ip>anchor) & (match3>lowMatchPtr)) && (ip[-1] == match3[-1])) { ip--; match3--; mLength++; } /* catch up */
} else {
const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend;
const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart;
mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4;
offset = curr - matchIndex;
while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
}
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
} else {
ip += ((ip-anchor) >> kSearchStrength) + 1;
continue;
} }
/* move to next sequence start */
ip += mLength;
anchor = ip;
if (ip <= ilimit) {
/* Complementary insertion */
/* done after iLimit test, as candidates could be > iend-8 */
{ U32 const indexToInsert = curr+2;
hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert;
hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert;
hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base);
}
/* check immediate repcode */
while (ip <= ilimit) {
U32 const current2 = (U32)(ip-base);
U32 const repIndex2 = current2 - offset_2;
const BYTE* repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) /* intentional overflow : ensure repIndex2 doesn't overlap dict + prefix */
& (offset_2 <= current2 - dictStartIndex))
&& (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
ip += repLength2;
anchor = ip;
continue;
}
break;
} } }
/* save reps for next block */
rep[0] = offset_1;
rep[1] = offset_2;
/* Return the last literals size */
return (size_t)(iend - anchor);
}
ZSTD_GEN_DFAST_FN(extDict, 4)
ZSTD_GEN_DFAST_FN(extDict, 5)
ZSTD_GEN_DFAST_FN(extDict, 6)
ZSTD_GEN_DFAST_FN(extDict, 7)
size_t ZSTD_compressBlock_doubleFast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_doubleFast_extDict_4(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_doubleFast_extDict_5(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_doubleFast_extDict_6(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_doubleFast_extDict_7(ms, seqStore, rep, src, srcSize);
}
}

39
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_double_fast.h vendored Normal file
View file

@ -0,0 +1,39 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_DOUBLE_FAST_H
#define ZSTD_DOUBLE_FAST_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "../common/mem.h" /* U32 */
#include "zstd_compress_internal.h" /* ZSTD_CCtx, size_t */
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp);
size_t ZSTD_compressBlock_doubleFast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_doubleFast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_doubleFast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_DOUBLE_FAST_H */

960
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_fast.c vendored Normal file
View file

@ -0,0 +1,960 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#include "zstd_compress_internal.h" /* ZSTD_hashPtr, ZSTD_count, ZSTD_storeSeq */
#include "zstd_fast.h"
static void ZSTD_fillHashTableForCDict(ZSTD_matchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hBits = cParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
U32 const mls = cParams->minMatch;
const BYTE* const base = ms->window.base;
const BYTE* ip = base + ms->nextToUpdate;
const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
const U32 fastHashFillStep = 3;
/* Currently, we always use ZSTD_dtlm_full for filling CDict tables.
* Feel free to remove this assert if there's a good reason! */
assert(dtlm == ZSTD_dtlm_full);
/* Always insert every fastHashFillStep position into the hash table.
* Insert the other positions if their hash entry is empty.
*/
for ( ; ip + fastHashFillStep < iend + 2; ip += fastHashFillStep) {
U32 const curr = (U32)(ip - base);
{ size_t const hashAndTag = ZSTD_hashPtr(ip, hBits, mls);
ZSTD_writeTaggedIndex(hashTable, hashAndTag, curr); }
if (dtlm == ZSTD_dtlm_fast) continue;
/* Only load extra positions for ZSTD_dtlm_full */
{ U32 p;
for (p = 1; p < fastHashFillStep; ++p) {
size_t const hashAndTag = ZSTD_hashPtr(ip + p, hBits, mls);
if (hashTable[hashAndTag >> ZSTD_SHORT_CACHE_TAG_BITS] == 0) { /* not yet filled */
ZSTD_writeTaggedIndex(hashTable, hashAndTag, curr + p);
} } } }
}
static void ZSTD_fillHashTableForCCtx(ZSTD_matchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hBits = cParams->hashLog;
U32 const mls = cParams->minMatch;
const BYTE* const base = ms->window.base;
const BYTE* ip = base + ms->nextToUpdate;
const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
const U32 fastHashFillStep = 3;
/* Currently, we always use ZSTD_dtlm_fast for filling CCtx tables.
* Feel free to remove this assert if there's a good reason! */
assert(dtlm == ZSTD_dtlm_fast);
/* Always insert every fastHashFillStep position into the hash table.
* Insert the other positions if their hash entry is empty.
*/
for ( ; ip + fastHashFillStep < iend + 2; ip += fastHashFillStep) {
U32 const curr = (U32)(ip - base);
size_t const hash0 = ZSTD_hashPtr(ip, hBits, mls);
hashTable[hash0] = curr;
if (dtlm == ZSTD_dtlm_fast) continue;
/* Only load extra positions for ZSTD_dtlm_full */
{ U32 p;
for (p = 1; p < fastHashFillStep; ++p) {
size_t const hash = ZSTD_hashPtr(ip + p, hBits, mls);
if (hashTable[hash] == 0) { /* not yet filled */
hashTable[hash] = curr + p;
} } } }
}
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp)
{
if (tfp == ZSTD_tfp_forCDict) {
ZSTD_fillHashTableForCDict(ms, end, dtlm);
} else {
ZSTD_fillHashTableForCCtx(ms, end, dtlm);
}
}
/**
* If you squint hard enough (and ignore repcodes), the search operation at any
* given position is broken into 4 stages:
*
* 1. Hash (map position to hash value via input read)
* 2. Lookup (map hash val to index via hashtable read)
* 3. Load (map index to value at that position via input read)
* 4. Compare
*
* Each of these steps involves a memory read at an address which is computed
* from the previous step. This means these steps must be sequenced and their
* latencies are cumulative.
*
* Rather than do 1->2->3->4 sequentially for a single position before moving
* onto the next, this implementation interleaves these operations across the
* next few positions:
*
* R = Repcode Read & Compare
* H = Hash
* T = Table Lookup
* M = Match Read & Compare
*
* Pos | Time -->
* ----+-------------------
* N | ... M
* N+1 | ... TM
* N+2 | R H T M
* N+3 | H TM
* N+4 | R H T M
* N+5 | H ...
* N+6 | R ...
*
* This is very much analogous to the pipelining of execution in a CPU. And just
* like a CPU, we have to dump the pipeline when we find a match (i.e., take a
* branch).
*
* When this happens, we throw away our current state, and do the following prep
* to re-enter the loop:
*
* Pos | Time -->
* ----+-------------------
* N | H T
* N+1 | H
*
* This is also the work we do at the beginning to enter the loop initially.
*/
FORCE_INLINE_TEMPLATE size_t
ZSTD_compressBlock_fast_noDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
U32 const mls, U32 const hasStep)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hlog = cParams->hashLog;
/* support stepSize of 0 */
size_t const stepSize = hasStep ? (cParams->targetLength + !(cParams->targetLength) + 1) : 2;
const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
const U32 prefixStartIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog);
const BYTE* const prefixStart = base + prefixStartIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
const BYTE* anchor = istart;
const BYTE* ip0 = istart;
const BYTE* ip1;
const BYTE* ip2;
const BYTE* ip3;
U32 current0;
U32 rep_offset1 = rep[0];
U32 rep_offset2 = rep[1];
U32 offsetSaved1 = 0, offsetSaved2 = 0;
size_t hash0; /* hash for ip0 */
size_t hash1; /* hash for ip1 */
U32 idx; /* match idx for ip0 */
U32 mval; /* src value at match idx */
U32 offcode;
const BYTE* match0;
size_t mLength;
/* ip0 and ip1 are always adjacent. The targetLength skipping and
* uncompressibility acceleration is applied to every other position,
* matching the behavior of #1562. step therefore represents the gap
* between pairs of positions, from ip0 to ip2 or ip1 to ip3. */
size_t step;
const BYTE* nextStep;
const size_t kStepIncr = (1 << (kSearchStrength - 1));
DEBUGLOG(5, "ZSTD_compressBlock_fast_generic");
ip0 += (ip0 == prefixStart);
{ U32 const curr = (U32)(ip0 - base);
U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, cParams->windowLog);
U32 const maxRep = curr - windowLow;
if (rep_offset2 > maxRep) offsetSaved2 = rep_offset2, rep_offset2 = 0;
if (rep_offset1 > maxRep) offsetSaved1 = rep_offset1, rep_offset1 = 0;
}
/* start each op */
_start: /* Requires: ip0 */
step = stepSize;
nextStep = ip0 + kStepIncr;
/* calculate positions, ip0 - anchor == 0, so we skip step calc */
ip1 = ip0 + 1;
ip2 = ip0 + step;
ip3 = ip2 + 1;
if (ip3 >= ilimit) {
goto _cleanup;
}
hash0 = ZSTD_hashPtr(ip0, hlog, mls);
hash1 = ZSTD_hashPtr(ip1, hlog, mls);
idx = hashTable[hash0];
do {
/* load repcode match for ip[2]*/
const U32 rval = MEM_read32(ip2 - rep_offset1);
/* write back hash table entry */
current0 = (U32)(ip0 - base);
hashTable[hash0] = current0;
/* check repcode at ip[2] */
if ((MEM_read32(ip2) == rval) & (rep_offset1 > 0)) {
ip0 = ip2;
match0 = ip0 - rep_offset1;
mLength = ip0[-1] == match0[-1];
ip0 -= mLength;
match0 -= mLength;
offcode = REPCODE1_TO_OFFBASE;
mLength += 4;
/* First write next hash table entry; we've already calculated it.
* This write is known to be safe because the ip1 is before the
* repcode (ip2). */
hashTable[hash1] = (U32)(ip1 - base);
goto _match;
}
/* load match for ip[0] */
if (idx >= prefixStartIndex) {
mval = MEM_read32(base + idx);
} else {
mval = MEM_read32(ip0) ^ 1; /* guaranteed to not match. */
}
/* check match at ip[0] */
if (MEM_read32(ip0) == mval) {
/* found a match! */
/* First write next hash table entry; we've already calculated it.
* This write is known to be safe because the ip1 == ip0 + 1, so
* we know we will resume searching after ip1 */
hashTable[hash1] = (U32)(ip1 - base);
goto _offset;
}
/* lookup ip[1] */
idx = hashTable[hash1];
/* hash ip[2] */
hash0 = hash1;
hash1 = ZSTD_hashPtr(ip2, hlog, mls);
/* advance to next positions */
ip0 = ip1;
ip1 = ip2;
ip2 = ip3;
/* write back hash table entry */
current0 = (U32)(ip0 - base);
hashTable[hash0] = current0;
/* load match for ip[0] */
if (idx >= prefixStartIndex) {
mval = MEM_read32(base + idx);
} else {
mval = MEM_read32(ip0) ^ 1; /* guaranteed to not match. */
}
/* check match at ip[0] */
if (MEM_read32(ip0) == mval) {
/* found a match! */
/* first write next hash table entry; we've already calculated it */
if (step <= 4) {
/* We need to avoid writing an index into the hash table >= the
* position at which we will pick up our searching after we've
* taken this match.
*
* The minimum possible match has length 4, so the earliest ip0
* can be after we take this match will be the current ip0 + 4.
* ip1 is ip0 + step - 1. If ip1 is >= ip0 + 4, we can't safely
* write this position.
*/
hashTable[hash1] = (U32)(ip1 - base);
}
goto _offset;
}
/* lookup ip[1] */
idx = hashTable[hash1];
/* hash ip[2] */
hash0 = hash1;
hash1 = ZSTD_hashPtr(ip2, hlog, mls);
/* advance to next positions */
ip0 = ip1;
ip1 = ip2;
ip2 = ip0 + step;
ip3 = ip1 + step;
/* calculate step */
if (ip2 >= nextStep) {
step++;
PREFETCH_L1(ip1 + 64);
PREFETCH_L1(ip1 + 128);
nextStep += kStepIncr;
}
} while (ip3 < ilimit);
_cleanup:
/* Note that there are probably still a couple positions we could search.
* However, it seems to be a meaningful performance hit to try to search
* them. So let's not. */
/* When the repcodes are outside of the prefix, we set them to zero before the loop.
* When the offsets are still zero, we need to restore them after the block to have a correct
* repcode history. If only one offset was invalid, it is easy. The tricky case is when both
* offsets were invalid. We need to figure out which offset to refill with.
* - If both offsets are zero they are in the same order.
* - If both offsets are non-zero, we won't restore the offsets from `offsetSaved[12]`.
* - If only one is zero, we need to decide which offset to restore.
* - If rep_offset1 is non-zero, then rep_offset2 must be offsetSaved1.
* - It is impossible for rep_offset2 to be non-zero.
*
* So if rep_offset1 started invalid (offsetSaved1 != 0) and became valid (rep_offset1 != 0), then
* set rep[0] = rep_offset1 and rep[1] = offsetSaved1.
*/
offsetSaved2 = ((offsetSaved1 != 0) && (rep_offset1 != 0)) ? offsetSaved1 : offsetSaved2;
/* save reps for next block */
rep[0] = rep_offset1 ? rep_offset1 : offsetSaved1;
rep[1] = rep_offset2 ? rep_offset2 : offsetSaved2;
/* Return the last literals size */
return (size_t)(iend - anchor);
_offset: /* Requires: ip0, idx */
/* Compute the offset code. */
match0 = base + idx;
rep_offset2 = rep_offset1;
rep_offset1 = (U32)(ip0-match0);
offcode = OFFSET_TO_OFFBASE(rep_offset1);
mLength = 4;
/* Count the backwards match length. */
while (((ip0>anchor) & (match0>prefixStart)) && (ip0[-1] == match0[-1])) {
ip0--;
match0--;
mLength++;
}
_match: /* Requires: ip0, match0, offcode */
/* Count the forward length. */
mLength += ZSTD_count(ip0 + mLength, match0 + mLength, iend);
ZSTD_storeSeq(seqStore, (size_t)(ip0 - anchor), anchor, iend, offcode, mLength);
ip0 += mLength;
anchor = ip0;
/* Fill table and check for immediate repcode. */
if (ip0 <= ilimit) {
/* Fill Table */
assert(base+current0+2 > istart); /* check base overflow */
hashTable[ZSTD_hashPtr(base+current0+2, hlog, mls)] = current0+2; /* here because current+2 could be > iend-8 */
hashTable[ZSTD_hashPtr(ip0-2, hlog, mls)] = (U32)(ip0-2-base);
if (rep_offset2 > 0) { /* rep_offset2==0 means rep_offset2 is invalidated */
while ( (ip0 <= ilimit) && (MEM_read32(ip0) == MEM_read32(ip0 - rep_offset2)) ) {
/* store sequence */
size_t const rLength = ZSTD_count(ip0+4, ip0+4-rep_offset2, iend) + 4;
{ U32 const tmpOff = rep_offset2; rep_offset2 = rep_offset1; rep_offset1 = tmpOff; } /* swap rep_offset2 <=> rep_offset1 */
hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = (U32)(ip0-base);
ip0 += rLength;
ZSTD_storeSeq(seqStore, 0 /*litLen*/, anchor, iend, REPCODE1_TO_OFFBASE, rLength);
anchor = ip0;
continue; /* faster when present (confirmed on gcc-8) ... (?) */
} } }
goto _start;
}
#define ZSTD_GEN_FAST_FN(dictMode, mls, step) \
static size_t ZSTD_compressBlock_fast_##dictMode##_##mls##_##step( \
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \
void const* src, size_t srcSize) \
{ \
return ZSTD_compressBlock_fast_##dictMode##_generic(ms, seqStore, rep, src, srcSize, mls, step); \
}
ZSTD_GEN_FAST_FN(noDict, 4, 1)
ZSTD_GEN_FAST_FN(noDict, 5, 1)
ZSTD_GEN_FAST_FN(noDict, 6, 1)
ZSTD_GEN_FAST_FN(noDict, 7, 1)
ZSTD_GEN_FAST_FN(noDict, 4, 0)
ZSTD_GEN_FAST_FN(noDict, 5, 0)
ZSTD_GEN_FAST_FN(noDict, 6, 0)
ZSTD_GEN_FAST_FN(noDict, 7, 0)
size_t ZSTD_compressBlock_fast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;
assert(ms->dictMatchState == NULL);
if (ms->cParams.targetLength > 1) {
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_fast_noDict_4_1(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_fast_noDict_5_1(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_fast_noDict_6_1(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_fast_noDict_7_1(ms, seqStore, rep, src, srcSize);
}
} else {
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_fast_noDict_4_0(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_fast_noDict_5_0(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_fast_noDict_6_0(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_fast_noDict_7_0(ms, seqStore, rep, src, srcSize);
}
}
}
FORCE_INLINE_TEMPLATE
size_t ZSTD_compressBlock_fast_dictMatchState_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize, U32 const mls, U32 const hasStep)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hlog = cParams->hashLog;
/* support stepSize of 0 */
U32 const stepSize = cParams->targetLength + !(cParams->targetLength);
const BYTE* const base = ms->window.base;
const BYTE* const istart = (const BYTE*)src;
const BYTE* ip0 = istart;
const BYTE* ip1 = ip0 + stepSize; /* we assert below that stepSize >= 1 */
const BYTE* anchor = istart;
const U32 prefixStartIndex = ms->window.dictLimit;
const BYTE* const prefixStart = base + prefixStartIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
const ZSTD_matchState_t* const dms = ms->dictMatchState;
const ZSTD_compressionParameters* const dictCParams = &dms->cParams ;
const U32* const dictHashTable = dms->hashTable;
const U32 dictStartIndex = dms->window.dictLimit;
const BYTE* const dictBase = dms->window.base;
const BYTE* const dictStart = dictBase + dictStartIndex;
const BYTE* const dictEnd = dms->window.nextSrc;
const U32 dictIndexDelta = prefixStartIndex - (U32)(dictEnd - dictBase);
const U32 dictAndPrefixLength = (U32)(istart - prefixStart + dictEnd - dictStart);
const U32 dictHBits = dictCParams->hashLog + ZSTD_SHORT_CACHE_TAG_BITS;
/* if a dictionary is still attached, it necessarily means that
* it is within window size. So we just check it. */
const U32 maxDistance = 1U << cParams->windowLog;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
assert(endIndex - prefixStartIndex <= maxDistance);
(void)maxDistance; (void)endIndex; /* these variables are not used when assert() is disabled */
(void)hasStep; /* not currently specialized on whether it's accelerated */
/* ensure there will be no underflow
* when translating a dict index into a local index */
assert(prefixStartIndex >= (U32)(dictEnd - dictBase));
if (ms->prefetchCDictTables) {
size_t const hashTableBytes = (((size_t)1) << dictCParams->hashLog) * sizeof(U32);
PREFETCH_AREA(dictHashTable, hashTableBytes)
}
/* init */
DEBUGLOG(5, "ZSTD_compressBlock_fast_dictMatchState_generic");
ip0 += (dictAndPrefixLength == 0);
/* dictMatchState repCode checks don't currently handle repCode == 0
* disabling. */
assert(offset_1 <= dictAndPrefixLength);
assert(offset_2 <= dictAndPrefixLength);
/* Outer search loop */
assert(stepSize >= 1);
while (ip1 <= ilimit) { /* repcode check at (ip0 + 1) is safe because ip0 < ip1 */
size_t mLength;
size_t hash0 = ZSTD_hashPtr(ip0, hlog, mls);
size_t const dictHashAndTag0 = ZSTD_hashPtr(ip0, dictHBits, mls);
U32 dictMatchIndexAndTag = dictHashTable[dictHashAndTag0 >> ZSTD_SHORT_CACHE_TAG_BITS];
int dictTagsMatch = ZSTD_comparePackedTags(dictMatchIndexAndTag, dictHashAndTag0);
U32 matchIndex = hashTable[hash0];
U32 curr = (U32)(ip0 - base);
size_t step = stepSize;
const size_t kStepIncr = 1 << kSearchStrength;
const BYTE* nextStep = ip0 + kStepIncr;
/* Inner search loop */
while (1) {
const BYTE* match = base + matchIndex;
const U32 repIndex = curr + 1 - offset_1;
const BYTE* repMatch = (repIndex < prefixStartIndex) ?
dictBase + (repIndex - dictIndexDelta) :
base + repIndex;
const size_t hash1 = ZSTD_hashPtr(ip1, hlog, mls);
size_t const dictHashAndTag1 = ZSTD_hashPtr(ip1, dictHBits, mls);
hashTable[hash0] = curr; /* update hash table */
if (((U32) ((prefixStartIndex - 1) - repIndex) >=
3) /* intentional underflow : ensure repIndex isn't overlapping dict + prefix */
&& (MEM_read32(repMatch) == MEM_read32(ip0 + 1))) {
const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
mLength = ZSTD_count_2segments(ip0 + 1 + 4, repMatch + 4, iend, repMatchEnd, prefixStart) + 4;
ip0++;
ZSTD_storeSeq(seqStore, (size_t) (ip0 - anchor), anchor, iend, REPCODE1_TO_OFFBASE, mLength);
break;
}
if (dictTagsMatch) {
/* Found a possible dict match */
const U32 dictMatchIndex = dictMatchIndexAndTag >> ZSTD_SHORT_CACHE_TAG_BITS;
const BYTE* dictMatch = dictBase + dictMatchIndex;
if (dictMatchIndex > dictStartIndex &&
MEM_read32(dictMatch) == MEM_read32(ip0)) {
/* To replicate extDict parse behavior, we only use dict matches when the normal matchIndex is invalid */
if (matchIndex <= prefixStartIndex) {
U32 const offset = (U32) (curr - dictMatchIndex - dictIndexDelta);
mLength = ZSTD_count_2segments(ip0 + 4, dictMatch + 4, iend, dictEnd, prefixStart) + 4;
while (((ip0 > anchor) & (dictMatch > dictStart))
&& (ip0[-1] == dictMatch[-1])) {
ip0--;
dictMatch--;
mLength++;
} /* catch up */
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t) (ip0 - anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
break;
}
}
}
if (matchIndex > prefixStartIndex && MEM_read32(match) == MEM_read32(ip0)) {
/* found a regular match */
U32 const offset = (U32) (ip0 - match);
mLength = ZSTD_count(ip0 + 4, match + 4, iend) + 4;
while (((ip0 > anchor) & (match > prefixStart))
&& (ip0[-1] == match[-1])) {
ip0--;
match--;
mLength++;
} /* catch up */
offset_2 = offset_1;
offset_1 = offset;
ZSTD_storeSeq(seqStore, (size_t) (ip0 - anchor), anchor, iend, OFFSET_TO_OFFBASE(offset), mLength);
break;
}
/* Prepare for next iteration */
dictMatchIndexAndTag = dictHashTable[dictHashAndTag1 >> ZSTD_SHORT_CACHE_TAG_BITS];
dictTagsMatch = ZSTD_comparePackedTags(dictMatchIndexAndTag, dictHashAndTag1);
matchIndex = hashTable[hash1];
if (ip1 >= nextStep) {
step++;
nextStep += kStepIncr;
}
ip0 = ip1;
ip1 = ip1 + step;
if (ip1 > ilimit) goto _cleanup;
curr = (U32)(ip0 - base);
hash0 = hash1;
} /* end inner search loop */
/* match found */
assert(mLength);
ip0 += mLength;
anchor = ip0;
if (ip0 <= ilimit) {
/* Fill Table */
assert(base+curr+2 > istart); /* check base overflow */
hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2; /* here because curr+2 could be > iend-8 */
hashTable[ZSTD_hashPtr(ip0-2, hlog, mls)] = (U32)(ip0-2-base);
/* check immediate repcode */
while (ip0 <= ilimit) {
U32 const current2 = (U32)(ip0-base);
U32 const repIndex2 = current2 - offset_2;
const BYTE* repMatch2 = repIndex2 < prefixStartIndex ?
dictBase - dictIndexDelta + repIndex2 :
base + repIndex2;
if ( ((U32)((prefixStartIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */)
&& (MEM_read32(repMatch2) == MEM_read32(ip0))) {
const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip0+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = current2;
ip0 += repLength2;
anchor = ip0;
continue;
}
break;
}
}
/* Prepare for next iteration */
assert(ip0 == anchor);
ip1 = ip0 + stepSize;
}
_cleanup:
/* save reps for next block */
rep[0] = offset_1;
rep[1] = offset_2;
/* Return the last literals size */
return (size_t)(iend - anchor);
}
ZSTD_GEN_FAST_FN(dictMatchState, 4, 0)
ZSTD_GEN_FAST_FN(dictMatchState, 5, 0)
ZSTD_GEN_FAST_FN(dictMatchState, 6, 0)
ZSTD_GEN_FAST_FN(dictMatchState, 7, 0)
size_t ZSTD_compressBlock_fast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;
assert(ms->dictMatchState != NULL);
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_fast_dictMatchState_4_0(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_fast_dictMatchState_5_0(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_fast_dictMatchState_6_0(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_fast_dictMatchState_7_0(ms, seqStore, rep, src, srcSize);
}
}
static size_t ZSTD_compressBlock_fast_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize, U32 const mls, U32 const hasStep)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
U32* const hashTable = ms->hashTable;
U32 const hlog = cParams->hashLog;
/* support stepSize of 0 */
size_t const stepSize = cParams->targetLength + !(cParams->targetLength) + 1;
const BYTE* const base = ms->window.base;
const BYTE* const dictBase = ms->window.dictBase;
const BYTE* const istart = (const BYTE*)src;
const BYTE* anchor = istart;
const U32 endIndex = (U32)((size_t)(istart - base) + srcSize);
const U32 lowLimit = ZSTD_getLowestMatchIndex(ms, endIndex, cParams->windowLog);
const U32 dictStartIndex = lowLimit;
const BYTE* const dictStart = dictBase + dictStartIndex;
const U32 dictLimit = ms->window.dictLimit;
const U32 prefixStartIndex = dictLimit < lowLimit ? lowLimit : dictLimit;
const BYTE* const prefixStart = base + prefixStartIndex;
const BYTE* const dictEnd = dictBase + prefixStartIndex;
const BYTE* const iend = istart + srcSize;
const BYTE* const ilimit = iend - 8;
U32 offset_1=rep[0], offset_2=rep[1];
U32 offsetSaved1 = 0, offsetSaved2 = 0;
const BYTE* ip0 = istart;
const BYTE* ip1;
const BYTE* ip2;
const BYTE* ip3;
U32 current0;
size_t hash0; /* hash for ip0 */
size_t hash1; /* hash for ip1 */
U32 idx; /* match idx for ip0 */
const BYTE* idxBase; /* base pointer for idx */
U32 offcode;
const BYTE* match0;
size_t mLength;
const BYTE* matchEnd = 0; /* initialize to avoid warning, assert != 0 later */
size_t step;
const BYTE* nextStep;
const size_t kStepIncr = (1 << (kSearchStrength - 1));
(void)hasStep; /* not currently specialized on whether it's accelerated */
DEBUGLOG(5, "ZSTD_compressBlock_fast_extDict_generic (offset_1=%u)", offset_1);
/* switch to "regular" variant if extDict is invalidated due to maxDistance */
if (prefixStartIndex == dictStartIndex)
return ZSTD_compressBlock_fast(ms, seqStore, rep, src, srcSize);
{ U32 const curr = (U32)(ip0 - base);
U32 const maxRep = curr - dictStartIndex;
if (offset_2 >= maxRep) offsetSaved2 = offset_2, offset_2 = 0;
if (offset_1 >= maxRep) offsetSaved1 = offset_1, offset_1 = 0;
}
/* start each op */
_start: /* Requires: ip0 */
step = stepSize;
nextStep = ip0 + kStepIncr;
/* calculate positions, ip0 - anchor == 0, so we skip step calc */
ip1 = ip0 + 1;
ip2 = ip0 + step;
ip3 = ip2 + 1;
if (ip3 >= ilimit) {
goto _cleanup;
}
hash0 = ZSTD_hashPtr(ip0, hlog, mls);
hash1 = ZSTD_hashPtr(ip1, hlog, mls);
idx = hashTable[hash0];
idxBase = idx < prefixStartIndex ? dictBase : base;
do {
{ /* load repcode match for ip[2] */
U32 const current2 = (U32)(ip2 - base);
U32 const repIndex = current2 - offset_1;
const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base;
U32 rval;
if ( ((U32)(prefixStartIndex - repIndex) >= 4) /* intentional underflow */
& (offset_1 > 0) ) {
rval = MEM_read32(repBase + repIndex);
} else {
rval = MEM_read32(ip2) ^ 1; /* guaranteed to not match. */
}
/* write back hash table entry */
current0 = (U32)(ip0 - base);
hashTable[hash0] = current0;
/* check repcode at ip[2] */
if (MEM_read32(ip2) == rval) {
ip0 = ip2;
match0 = repBase + repIndex;
matchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
assert((match0 != prefixStart) & (match0 != dictStart));
mLength = ip0[-1] == match0[-1];
ip0 -= mLength;
match0 -= mLength;
offcode = REPCODE1_TO_OFFBASE;
mLength += 4;
goto _match;
} }
{ /* load match for ip[0] */
U32 const mval = idx >= dictStartIndex ?
MEM_read32(idxBase + idx) :
MEM_read32(ip0) ^ 1; /* guaranteed not to match */
/* check match at ip[0] */
if (MEM_read32(ip0) == mval) {
/* found a match! */
goto _offset;
} }
/* lookup ip[1] */
idx = hashTable[hash1];
idxBase = idx < prefixStartIndex ? dictBase : base;
/* hash ip[2] */
hash0 = hash1;
hash1 = ZSTD_hashPtr(ip2, hlog, mls);
/* advance to next positions */
ip0 = ip1;
ip1 = ip2;
ip2 = ip3;
/* write back hash table entry */
current0 = (U32)(ip0 - base);
hashTable[hash0] = current0;
{ /* load match for ip[0] */
U32 const mval = idx >= dictStartIndex ?
MEM_read32(idxBase + idx) :
MEM_read32(ip0) ^ 1; /* guaranteed not to match */
/* check match at ip[0] */
if (MEM_read32(ip0) == mval) {
/* found a match! */
goto _offset;
} }
/* lookup ip[1] */
idx = hashTable[hash1];
idxBase = idx < prefixStartIndex ? dictBase : base;
/* hash ip[2] */
hash0 = hash1;
hash1 = ZSTD_hashPtr(ip2, hlog, mls);
/* advance to next positions */
ip0 = ip1;
ip1 = ip2;
ip2 = ip0 + step;
ip3 = ip1 + step;
/* calculate step */
if (ip2 >= nextStep) {
step++;
PREFETCH_L1(ip1 + 64);
PREFETCH_L1(ip1 + 128);
nextStep += kStepIncr;
}
} while (ip3 < ilimit);
_cleanup:
/* Note that there are probably still a couple positions we could search.
* However, it seems to be a meaningful performance hit to try to search
* them. So let's not. */
/* If offset_1 started invalid (offsetSaved1 != 0) and became valid (offset_1 != 0),
* rotate saved offsets. See comment in ZSTD_compressBlock_fast_noDict for more context. */
offsetSaved2 = ((offsetSaved1 != 0) && (offset_1 != 0)) ? offsetSaved1 : offsetSaved2;
/* save reps for next block */
rep[0] = offset_1 ? offset_1 : offsetSaved1;
rep[1] = offset_2 ? offset_2 : offsetSaved2;
/* Return the last literals size */
return (size_t)(iend - anchor);
_offset: /* Requires: ip0, idx, idxBase */
/* Compute the offset code. */
{ U32 const offset = current0 - idx;
const BYTE* const lowMatchPtr = idx < prefixStartIndex ? dictStart : prefixStart;
matchEnd = idx < prefixStartIndex ? dictEnd : iend;
match0 = idxBase + idx;
offset_2 = offset_1;
offset_1 = offset;
offcode = OFFSET_TO_OFFBASE(offset);
mLength = 4;
/* Count the backwards match length. */
while (((ip0>anchor) & (match0>lowMatchPtr)) && (ip0[-1] == match0[-1])) {
ip0--;
match0--;
mLength++;
} }
_match: /* Requires: ip0, match0, offcode, matchEnd */
/* Count the forward length. */
assert(matchEnd != 0);
mLength += ZSTD_count_2segments(ip0 + mLength, match0 + mLength, iend, matchEnd, prefixStart);
ZSTD_storeSeq(seqStore, (size_t)(ip0 - anchor), anchor, iend, offcode, mLength);
ip0 += mLength;
anchor = ip0;
/* write next hash table entry */
if (ip1 < ip0) {
hashTable[hash1] = (U32)(ip1 - base);
}
/* Fill table and check for immediate repcode. */
if (ip0 <= ilimit) {
/* Fill Table */
assert(base+current0+2 > istart); /* check base overflow */
hashTable[ZSTD_hashPtr(base+current0+2, hlog, mls)] = current0+2; /* here because current+2 could be > iend-8 */
hashTable[ZSTD_hashPtr(ip0-2, hlog, mls)] = (U32)(ip0-2-base);
while (ip0 <= ilimit) {
U32 const repIndex2 = (U32)(ip0-base) - offset_2;
const BYTE* const repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2;
if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) & (offset_2 > 0)) /* intentional underflow */
&& (MEM_read32(repMatch2) == MEM_read32(ip0)) ) {
const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip0+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
{ U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; } /* swap offset_2 <=> offset_1 */
ZSTD_storeSeq(seqStore, 0 /*litlen*/, anchor, iend, REPCODE1_TO_OFFBASE, repLength2);
hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = (U32)(ip0-base);
ip0 += repLength2;
anchor = ip0;
continue;
}
break;
} }
goto _start;
}
ZSTD_GEN_FAST_FN(extDict, 4, 0)
ZSTD_GEN_FAST_FN(extDict, 5, 0)
ZSTD_GEN_FAST_FN(extDict, 6, 0)
ZSTD_GEN_FAST_FN(extDict, 7, 0)
size_t ZSTD_compressBlock_fast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;
assert(ms->dictMatchState == NULL);
switch(mls)
{
default: /* includes case 3 */
case 4 :
return ZSTD_compressBlock_fast_extDict_4_0(ms, seqStore, rep, src, srcSize);
case 5 :
return ZSTD_compressBlock_fast_extDict_5_0(ms, seqStore, rep, src, srcSize);
case 6 :
return ZSTD_compressBlock_fast_extDict_6_0(ms, seqStore, rep, src, srcSize);
case 7 :
return ZSTD_compressBlock_fast_extDict_7_0(ms, seqStore, rep, src, srcSize);
}
}

38
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_fast.h vendored Normal file
View file

@ -0,0 +1,38 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_FAST_H
#define ZSTD_FAST_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "../common/mem.h" /* U32 */
#include "zstd_compress_internal.h"
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp);
size_t ZSTD_compressBlock_fast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_fast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_fast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_FAST_H */

2157
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_lazy.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

127
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_lazy.h vendored Normal file
View file

@ -0,0 +1,127 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_LAZY_H
#define ZSTD_LAZY_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "zstd_compress_internal.h"
/**
* Dedicated Dictionary Search Structure bucket log. In the
* ZSTD_dedicatedDictSearch mode, the hashTable has
* 2 ** ZSTD_LAZY_DDSS_BUCKET_LOG entries in each bucket, rather than just
* one.
*/
#define ZSTD_LAZY_DDSS_BUCKET_LOG 2
#define ZSTD_ROW_HASH_TAG_BITS 8 /* nb bits to use for the tag */
U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip);
void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip);
void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip);
void ZSTD_preserveUnsortedMark (U32* const table, U32 const size, U32 const reducerValue); /*! used in ZSTD_reduceIndex(). preemptively increase value of ZSTD_DUBT_UNSORTED_MARK */
size_t ZSTD_compressBlock_btlazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btlazy2_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btlazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_LAZY_H */

724
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_ldm.c vendored Normal file
View file

@ -0,0 +1,724 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#include "zstd_ldm.h"
#include "../common/debug.h"
#include "../common/xxhash.h"
#include "zstd_fast.h" /* ZSTD_fillHashTable() */
#include "zstd_double_fast.h" /* ZSTD_fillDoubleHashTable() */
#include "zstd_ldm_geartab.h"
#define LDM_BUCKET_SIZE_LOG 3
#define LDM_MIN_MATCH_LENGTH 64
#define LDM_HASH_RLOG 7
typedef struct {
U64 rolling;
U64 stopMask;
} ldmRollingHashState_t;
/** ZSTD_ldm_gear_init():
*
* Initializes the rolling hash state such that it will honor the
* settings in params. */
static void ZSTD_ldm_gear_init(ldmRollingHashState_t* state, ldmParams_t const* params)
{
unsigned maxBitsInMask = MIN(params->minMatchLength, 64);
unsigned hashRateLog = params->hashRateLog;
state->rolling = ~(U32)0;
/* The choice of the splitting criterion is subject to two conditions:
* 1. it has to trigger on average every 2^(hashRateLog) bytes;
* 2. ideally, it has to depend on a window of minMatchLength bytes.
*
* In the gear hash algorithm, bit n depends on the last n bytes;
* so in order to obtain a good quality splitting criterion it is
* preferable to use bits with high weight.
*
* To match condition 1 we use a mask with hashRateLog bits set
* and, because of the previous remark, we make sure these bits
* have the highest possible weight while still respecting
* condition 2.
*/
if (hashRateLog > 0 && hashRateLog <= maxBitsInMask) {
state->stopMask = (((U64)1 << hashRateLog) - 1) << (maxBitsInMask - hashRateLog);
} else {
/* In this degenerate case we simply honor the hash rate. */
state->stopMask = ((U64)1 << hashRateLog) - 1;
}
}
/** ZSTD_ldm_gear_reset()
* Feeds [data, data + minMatchLength) into the hash without registering any
* splits. This effectively resets the hash state. This is used when skipping
* over data, either at the beginning of a block, or skipping sections.
*/
static void ZSTD_ldm_gear_reset(ldmRollingHashState_t* state,
BYTE const* data, size_t minMatchLength)
{
U64 hash = state->rolling;
size_t n = 0;
#define GEAR_ITER_ONCE() do { \
hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \
n += 1; \
} while (0)
while (n + 3 < minMatchLength) {
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
}
while (n < minMatchLength) {
GEAR_ITER_ONCE();
}
#undef GEAR_ITER_ONCE
}
/** ZSTD_ldm_gear_feed():
*
* Registers in the splits array all the split points found in the first
* size bytes following the data pointer. This function terminates when
* either all the data has been processed or LDM_BATCH_SIZE splits are
* present in the splits array.
*
* Precondition: The splits array must not be full.
* Returns: The number of bytes processed. */
static size_t ZSTD_ldm_gear_feed(ldmRollingHashState_t* state,
BYTE const* data, size_t size,
size_t* splits, unsigned* numSplits)
{
size_t n;
U64 hash, mask;
hash = state->rolling;
mask = state->stopMask;
n = 0;
#define GEAR_ITER_ONCE() do { \
hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \
n += 1; \
if (UNLIKELY((hash & mask) == 0)) { \
splits[*numSplits] = n; \
*numSplits += 1; \
if (*numSplits == LDM_BATCH_SIZE) \
goto done; \
} \
} while (0)
while (n + 3 < size) {
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
GEAR_ITER_ONCE();
}
while (n < size) {
GEAR_ITER_ONCE();
}
#undef GEAR_ITER_ONCE
done:
state->rolling = hash;
return n;
}
void ZSTD_ldm_adjustParameters(ldmParams_t* params,
ZSTD_compressionParameters const* cParams)
{
params->windowLog = cParams->windowLog;
ZSTD_STATIC_ASSERT(LDM_BUCKET_SIZE_LOG <= ZSTD_LDM_BUCKETSIZELOG_MAX);
DEBUGLOG(4, "ZSTD_ldm_adjustParameters");
if (!params->bucketSizeLog) params->bucketSizeLog = LDM_BUCKET_SIZE_LOG;
if (!params->minMatchLength) params->minMatchLength = LDM_MIN_MATCH_LENGTH;
if (params->hashLog == 0) {
params->hashLog = MAX(ZSTD_HASHLOG_MIN, params->windowLog - LDM_HASH_RLOG);
assert(params->hashLog <= ZSTD_HASHLOG_MAX);
}
if (params->hashRateLog == 0) {
params->hashRateLog = params->windowLog < params->hashLog
? 0
: params->windowLog - params->hashLog;
}
params->bucketSizeLog = MIN(params->bucketSizeLog, params->hashLog);
}
size_t ZSTD_ldm_getTableSize(ldmParams_t params)
{
size_t const ldmHSize = ((size_t)1) << params.hashLog;
size_t const ldmBucketSizeLog = MIN(params.bucketSizeLog, params.hashLog);
size_t const ldmBucketSize = ((size_t)1) << (params.hashLog - ldmBucketSizeLog);
size_t const totalSize = ZSTD_cwksp_alloc_size(ldmBucketSize)
+ ZSTD_cwksp_alloc_size(ldmHSize * sizeof(ldmEntry_t));
return params.enableLdm == ZSTD_ps_enable ? totalSize : 0;
}
size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize)
{
return params.enableLdm == ZSTD_ps_enable ? (maxChunkSize / params.minMatchLength) : 0;
}
/** ZSTD_ldm_getBucket() :
* Returns a pointer to the start of the bucket associated with hash. */
static ldmEntry_t* ZSTD_ldm_getBucket(
ldmState_t* ldmState, size_t hash, ldmParams_t const ldmParams)
{
return ldmState->hashTable + (hash << ldmParams.bucketSizeLog);
}
/** ZSTD_ldm_insertEntry() :
* Insert the entry with corresponding hash into the hash table */
static void ZSTD_ldm_insertEntry(ldmState_t* ldmState,
size_t const hash, const ldmEntry_t entry,
ldmParams_t const ldmParams)
{
BYTE* const pOffset = ldmState->bucketOffsets + hash;
unsigned const offset = *pOffset;
*(ZSTD_ldm_getBucket(ldmState, hash, ldmParams) + offset) = entry;
*pOffset = (BYTE)((offset + 1) & ((1u << ldmParams.bucketSizeLog) - 1));
}
/** ZSTD_ldm_countBackwardsMatch() :
* Returns the number of bytes that match backwards before pIn and pMatch.
*
* We count only bytes where pMatch >= pBase and pIn >= pAnchor. */
static size_t ZSTD_ldm_countBackwardsMatch(
const BYTE* pIn, const BYTE* pAnchor,
const BYTE* pMatch, const BYTE* pMatchBase)
{
size_t matchLength = 0;
while (pIn > pAnchor && pMatch > pMatchBase && pIn[-1] == pMatch[-1]) {
pIn--;
pMatch--;
matchLength++;
}
return matchLength;
}
/** ZSTD_ldm_countBackwardsMatch_2segments() :
* Returns the number of bytes that match backwards from pMatch,
* even with the backwards match spanning 2 different segments.
*
* On reaching `pMatchBase`, start counting from mEnd */
static size_t ZSTD_ldm_countBackwardsMatch_2segments(
const BYTE* pIn, const BYTE* pAnchor,
const BYTE* pMatch, const BYTE* pMatchBase,
const BYTE* pExtDictStart, const BYTE* pExtDictEnd)
{
size_t matchLength = ZSTD_ldm_countBackwardsMatch(pIn, pAnchor, pMatch, pMatchBase);
if (pMatch - matchLength != pMatchBase || pMatchBase == pExtDictStart) {
/* If backwards match is entirely in the extDict or prefix, immediately return */
return matchLength;
}
DEBUGLOG(7, "ZSTD_ldm_countBackwardsMatch_2segments: found 2-parts backwards match (length in prefix==%zu)", matchLength);
matchLength += ZSTD_ldm_countBackwardsMatch(pIn - matchLength, pAnchor, pExtDictEnd, pExtDictStart);
DEBUGLOG(7, "final backwards match length = %zu", matchLength);
return matchLength;
}
/** ZSTD_ldm_fillFastTables() :
*
* Fills the relevant tables for the ZSTD_fast and ZSTD_dfast strategies.
* This is similar to ZSTD_loadDictionaryContent.
*
* The tables for the other strategies are filled within their
* block compressors. */
static size_t ZSTD_ldm_fillFastTables(ZSTD_matchState_t* ms,
void const* end)
{
const BYTE* const iend = (const BYTE*)end;
switch(ms->cParams.strategy)
{
case ZSTD_fast:
ZSTD_fillHashTable(ms, iend, ZSTD_dtlm_fast, ZSTD_tfp_forCCtx);
break;
case ZSTD_dfast:
ZSTD_fillDoubleHashTable(ms, iend, ZSTD_dtlm_fast, ZSTD_tfp_forCCtx);
break;
case ZSTD_greedy:
case ZSTD_lazy:
case ZSTD_lazy2:
case ZSTD_btlazy2:
case ZSTD_btopt:
case ZSTD_btultra:
case ZSTD_btultra2:
break;
default:
assert(0); /* not possible : not a valid strategy id */
}
return 0;
}
void ZSTD_ldm_fillHashTable(
ldmState_t* ldmState, const BYTE* ip,
const BYTE* iend, ldmParams_t const* params)
{
U32 const minMatchLength = params->minMatchLength;
U32 const hBits = params->hashLog - params->bucketSizeLog;
BYTE const* const base = ldmState->window.base;
BYTE const* const istart = ip;
ldmRollingHashState_t hashState;
size_t* const splits = ldmState->splitIndices;
unsigned numSplits;
DEBUGLOG(5, "ZSTD_ldm_fillHashTable");
ZSTD_ldm_gear_init(&hashState, params);
while (ip < iend) {
size_t hashed;
unsigned n;
numSplits = 0;
hashed = ZSTD_ldm_gear_feed(&hashState, ip, iend - ip, splits, &numSplits);
for (n = 0; n < numSplits; n++) {
if (ip + splits[n] >= istart + minMatchLength) {
BYTE const* const split = ip + splits[n] - minMatchLength;
U64 const xxhash = XXH64(split, minMatchLength, 0);
U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1));
ldmEntry_t entry;
entry.offset = (U32)(split - base);
entry.checksum = (U32)(xxhash >> 32);
ZSTD_ldm_insertEntry(ldmState, hash, entry, *params);
}
}
ip += hashed;
}
}
/** ZSTD_ldm_limitTableUpdate() :
*
* Sets cctx->nextToUpdate to a position corresponding closer to anchor
* if it is far way
* (after a long match, only update tables a limited amount). */
static void ZSTD_ldm_limitTableUpdate(ZSTD_matchState_t* ms, const BYTE* anchor)
{
U32 const curr = (U32)(anchor - ms->window.base);
if (curr > ms->nextToUpdate + 1024) {
ms->nextToUpdate =
curr - MIN(512, curr - ms->nextToUpdate - 1024);
}
}
static size_t ZSTD_ldm_generateSequences_internal(
ldmState_t* ldmState, rawSeqStore_t* rawSeqStore,
ldmParams_t const* params, void const* src, size_t srcSize)
{
/* LDM parameters */
int const extDict = ZSTD_window_hasExtDict(ldmState->window);
U32 const minMatchLength = params->minMatchLength;
U32 const entsPerBucket = 1U << params->bucketSizeLog;
U32 const hBits = params->hashLog - params->bucketSizeLog;
/* Prefix and extDict parameters */
U32 const dictLimit = ldmState->window.dictLimit;
U32 const lowestIndex = extDict ? ldmState->window.lowLimit : dictLimit;
BYTE const* const base = ldmState->window.base;
BYTE const* const dictBase = extDict ? ldmState->window.dictBase : NULL;
BYTE const* const dictStart = extDict ? dictBase + lowestIndex : NULL;
BYTE const* const dictEnd = extDict ? dictBase + dictLimit : NULL;
BYTE const* const lowPrefixPtr = base + dictLimit;
/* Input bounds */
BYTE const* const istart = (BYTE const*)src;
BYTE const* const iend = istart + srcSize;
BYTE const* const ilimit = iend - HASH_READ_SIZE;
/* Input positions */
BYTE const* anchor = istart;
BYTE const* ip = istart;
/* Rolling hash state */
ldmRollingHashState_t hashState;
/* Arrays for staged-processing */
size_t* const splits = ldmState->splitIndices;
ldmMatchCandidate_t* const candidates = ldmState->matchCandidates;
unsigned numSplits;
if (srcSize < minMatchLength)
return iend - anchor;
/* Initialize the rolling hash state with the first minMatchLength bytes */
ZSTD_ldm_gear_init(&hashState, params);
ZSTD_ldm_gear_reset(&hashState, ip, minMatchLength);
ip += minMatchLength;
while (ip < ilimit) {
size_t hashed;
unsigned n;
numSplits = 0;
hashed = ZSTD_ldm_gear_feed(&hashState, ip, ilimit - ip,
splits, &numSplits);
for (n = 0; n < numSplits; n++) {
BYTE const* const split = ip + splits[n] - minMatchLength;
U64 const xxhash = XXH64(split, minMatchLength, 0);
U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1));
candidates[n].split = split;
candidates[n].hash = hash;
candidates[n].checksum = (U32)(xxhash >> 32);
candidates[n].bucket = ZSTD_ldm_getBucket(ldmState, hash, *params);
PREFETCH_L1(candidates[n].bucket);
}
for (n = 0; n < numSplits; n++) {
size_t forwardMatchLength = 0, backwardMatchLength = 0,
bestMatchLength = 0, mLength;
U32 offset;
BYTE const* const split = candidates[n].split;
U32 const checksum = candidates[n].checksum;
U32 const hash = candidates[n].hash;
ldmEntry_t* const bucket = candidates[n].bucket;
ldmEntry_t const* cur;
ldmEntry_t const* bestEntry = NULL;
ldmEntry_t newEntry;
newEntry.offset = (U32)(split - base);
newEntry.checksum = checksum;
/* If a split point would generate a sequence overlapping with
* the previous one, we merely register it in the hash table and
* move on */
if (split < anchor) {
ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params);
continue;
}
for (cur = bucket; cur < bucket + entsPerBucket; cur++) {
size_t curForwardMatchLength, curBackwardMatchLength,
curTotalMatchLength;
if (cur->checksum != checksum || cur->offset <= lowestIndex) {
continue;
}
if (extDict) {
BYTE const* const curMatchBase =
cur->offset < dictLimit ? dictBase : base;
BYTE const* const pMatch = curMatchBase + cur->offset;
BYTE const* const matchEnd =
cur->offset < dictLimit ? dictEnd : iend;
BYTE const* const lowMatchPtr =
cur->offset < dictLimit ? dictStart : lowPrefixPtr;
curForwardMatchLength =
ZSTD_count_2segments(split, pMatch, iend, matchEnd, lowPrefixPtr);
if (curForwardMatchLength < minMatchLength) {
continue;
}
curBackwardMatchLength = ZSTD_ldm_countBackwardsMatch_2segments(
split, anchor, pMatch, lowMatchPtr, dictStart, dictEnd);
} else { /* !extDict */
BYTE const* const pMatch = base + cur->offset;
curForwardMatchLength = ZSTD_count(split, pMatch, iend);
if (curForwardMatchLength < minMatchLength) {
continue;
}
curBackwardMatchLength =
ZSTD_ldm_countBackwardsMatch(split, anchor, pMatch, lowPrefixPtr);
}
curTotalMatchLength = curForwardMatchLength + curBackwardMatchLength;
if (curTotalMatchLength > bestMatchLength) {
bestMatchLength = curTotalMatchLength;
forwardMatchLength = curForwardMatchLength;
backwardMatchLength = curBackwardMatchLength;
bestEntry = cur;
}
}
/* No match found -- insert an entry into the hash table
* and process the next candidate match */
if (bestEntry == NULL) {
ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params);
continue;
}
/* Match found */
offset = (U32)(split - base) - bestEntry->offset;
mLength = forwardMatchLength + backwardMatchLength;
{
rawSeq* const seq = rawSeqStore->seq + rawSeqStore->size;
/* Out of sequence storage */
if (rawSeqStore->size == rawSeqStore->capacity)
return ERROR(dstSize_tooSmall);
seq->litLength = (U32)(split - backwardMatchLength - anchor);
seq->matchLength = (U32)mLength;
seq->offset = offset;
rawSeqStore->size++;
}
/* Insert the current entry into the hash table --- it must be
* done after the previous block to avoid clobbering bestEntry */
ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params);
anchor = split + forwardMatchLength;
/* If we find a match that ends after the data that we've hashed
* then we have a repeating, overlapping, pattern. E.g. all zeros.
* If one repetition of the pattern matches our `stopMask` then all
* repetitions will. We don't need to insert them all into out table,
* only the first one. So skip over overlapping matches.
* This is a major speed boost (20x) for compressing a single byte
* repeated, when that byte ends up in the table.
*/
if (anchor > ip + hashed) {
ZSTD_ldm_gear_reset(&hashState, anchor - minMatchLength, minMatchLength);
/* Continue the outer loop at anchor (ip + hashed == anchor). */
ip = anchor - hashed;
break;
}
}
ip += hashed;
}
return iend - anchor;
}
/*! ZSTD_ldm_reduceTable() :
* reduce table indexes by `reducerValue` */
static void ZSTD_ldm_reduceTable(ldmEntry_t* const table, U32 const size,
U32 const reducerValue)
{
U32 u;
for (u = 0; u < size; u++) {
if (table[u].offset < reducerValue) table[u].offset = 0;
else table[u].offset -= reducerValue;
}
}
size_t ZSTD_ldm_generateSequences(
ldmState_t* ldmState, rawSeqStore_t* sequences,
ldmParams_t const* params, void const* src, size_t srcSize)
{
U32 const maxDist = 1U << params->windowLog;
BYTE const* const istart = (BYTE const*)src;
BYTE const* const iend = istart + srcSize;
size_t const kMaxChunkSize = 1 << 20;
size_t const nbChunks = (srcSize / kMaxChunkSize) + ((srcSize % kMaxChunkSize) != 0);
size_t chunk;
size_t leftoverSize = 0;
assert(ZSTD_CHUNKSIZE_MAX >= kMaxChunkSize);
/* Check that ZSTD_window_update() has been called for this chunk prior
* to passing it to this function.
*/
assert(ldmState->window.nextSrc >= (BYTE const*)src + srcSize);
/* The input could be very large (in zstdmt), so it must be broken up into
* chunks to enforce the maximum distance and handle overflow correction.
*/
assert(sequences->pos <= sequences->size);
assert(sequences->size <= sequences->capacity);
for (chunk = 0; chunk < nbChunks && sequences->size < sequences->capacity; ++chunk) {
BYTE const* const chunkStart = istart + chunk * kMaxChunkSize;
size_t const remaining = (size_t)(iend - chunkStart);
BYTE const *const chunkEnd =
(remaining < kMaxChunkSize) ? iend : chunkStart + kMaxChunkSize;
size_t const chunkSize = chunkEnd - chunkStart;
size_t newLeftoverSize;
size_t const prevSize = sequences->size;
assert(chunkStart < iend);
/* 1. Perform overflow correction if necessary. */
if (ZSTD_window_needOverflowCorrection(ldmState->window, 0, maxDist, ldmState->loadedDictEnd, chunkStart, chunkEnd)) {
U32 const ldmHSize = 1U << params->hashLog;
U32 const correction = ZSTD_window_correctOverflow(
&ldmState->window, /* cycleLog */ 0, maxDist, chunkStart);
ZSTD_ldm_reduceTable(ldmState->hashTable, ldmHSize, correction);
/* invalidate dictionaries on overflow correction */
ldmState->loadedDictEnd = 0;
}
/* 2. We enforce the maximum offset allowed.
*
* kMaxChunkSize should be small enough that we don't lose too much of
* the window through early invalidation.
* TODO: * Test the chunk size.
* * Try invalidation after the sequence generation and test the
* offset against maxDist directly.
*
* NOTE: Because of dictionaries + sequence splitting we MUST make sure
* that any offset used is valid at the END of the sequence, since it may
* be split into two sequences. This condition holds when using
* ZSTD_window_enforceMaxDist(), but if we move to checking offsets
* against maxDist directly, we'll have to carefully handle that case.
*/
ZSTD_window_enforceMaxDist(&ldmState->window, chunkEnd, maxDist, &ldmState->loadedDictEnd, NULL);
/* 3. Generate the sequences for the chunk, and get newLeftoverSize. */
newLeftoverSize = ZSTD_ldm_generateSequences_internal(
ldmState, sequences, params, chunkStart, chunkSize);
if (ZSTD_isError(newLeftoverSize))
return newLeftoverSize;
/* 4. We add the leftover literals from previous iterations to the first
* newly generated sequence, or add the `newLeftoverSize` if none are
* generated.
*/
/* Prepend the leftover literals from the last call */
if (prevSize < sequences->size) {
sequences->seq[prevSize].litLength += (U32)leftoverSize;
leftoverSize = newLeftoverSize;
} else {
assert(newLeftoverSize == chunkSize);
leftoverSize += chunkSize;
}
}
return 0;
}
void
ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const minMatch)
{
while (srcSize > 0 && rawSeqStore->pos < rawSeqStore->size) {
rawSeq* seq = rawSeqStore->seq + rawSeqStore->pos;
if (srcSize <= seq->litLength) {
/* Skip past srcSize literals */
seq->litLength -= (U32)srcSize;
return;
}
srcSize -= seq->litLength;
seq->litLength = 0;
if (srcSize < seq->matchLength) {
/* Skip past the first srcSize of the match */
seq->matchLength -= (U32)srcSize;
if (seq->matchLength < minMatch) {
/* The match is too short, omit it */
if (rawSeqStore->pos + 1 < rawSeqStore->size) {
seq[1].litLength += seq[0].matchLength;
}
rawSeqStore->pos++;
}
return;
}
srcSize -= seq->matchLength;
seq->matchLength = 0;
rawSeqStore->pos++;
}
}
/**
* If the sequence length is longer than remaining then the sequence is split
* between this block and the next.
*
* Returns the current sequence to handle, or if the rest of the block should
* be literals, it returns a sequence with offset == 0.
*/
static rawSeq maybeSplitSequence(rawSeqStore_t* rawSeqStore,
U32 const remaining, U32 const minMatch)
{
rawSeq sequence = rawSeqStore->seq[rawSeqStore->pos];
assert(sequence.offset > 0);
/* Likely: No partial sequence */
if (remaining >= sequence.litLength + sequence.matchLength) {
rawSeqStore->pos++;
return sequence;
}
/* Cut the sequence short (offset == 0 ==> rest is literals). */
if (remaining <= sequence.litLength) {
sequence.offset = 0;
} else if (remaining < sequence.litLength + sequence.matchLength) {
sequence.matchLength = remaining - sequence.litLength;
if (sequence.matchLength < minMatch) {
sequence.offset = 0;
}
}
/* Skip past `remaining` bytes for the future sequences. */
ZSTD_ldm_skipSequences(rawSeqStore, remaining, minMatch);
return sequence;
}
void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) {
U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes);
while (currPos && rawSeqStore->pos < rawSeqStore->size) {
rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos];
if (currPos >= currSeq.litLength + currSeq.matchLength) {
currPos -= currSeq.litLength + currSeq.matchLength;
rawSeqStore->pos++;
} else {
rawSeqStore->posInSequence = currPos;
break;
}
}
if (currPos == 0 || rawSeqStore->pos == rawSeqStore->size) {
rawSeqStore->posInSequence = 0;
}
}
size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_paramSwitch_e useRowMatchFinder,
void const* src, size_t srcSize)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
unsigned const minMatch = cParams->minMatch;
ZSTD_blockCompressor const blockCompressor =
ZSTD_selectBlockCompressor(cParams->strategy, useRowMatchFinder, ZSTD_matchState_dictMode(ms));
/* Input bounds */
BYTE const* const istart = (BYTE const*)src;
BYTE const* const iend = istart + srcSize;
/* Input positions */
BYTE const* ip = istart;
DEBUGLOG(5, "ZSTD_ldm_blockCompress: srcSize=%zu", srcSize);
/* If using opt parser, use LDMs only as candidates rather than always accepting them */
if (cParams->strategy >= ZSTD_btopt) {
size_t lastLLSize;
ms->ldmSeqStore = rawSeqStore;
lastLLSize = blockCompressor(ms, seqStore, rep, src, srcSize);
ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore, srcSize);
return lastLLSize;
}
assert(rawSeqStore->pos <= rawSeqStore->size);
assert(rawSeqStore->size <= rawSeqStore->capacity);
/* Loop through each sequence and apply the block compressor to the literals */
while (rawSeqStore->pos < rawSeqStore->size && ip < iend) {
/* maybeSplitSequence updates rawSeqStore->pos */
rawSeq const sequence = maybeSplitSequence(rawSeqStore,
(U32)(iend - ip), minMatch);
int i;
/* End signal */
if (sequence.offset == 0)
break;
assert(ip + sequence.litLength + sequence.matchLength <= iend);
/* Fill tables for block compressor */
ZSTD_ldm_limitTableUpdate(ms, ip);
ZSTD_ldm_fillFastTables(ms, ip);
/* Run the block compressor */
DEBUGLOG(5, "pos %u : calling block compressor on segment of size %u", (unsigned)(ip-istart), sequence.litLength);
{
size_t const newLitLength =
blockCompressor(ms, seqStore, rep, ip, sequence.litLength);
ip += sequence.litLength;
/* Update the repcodes */
for (i = ZSTD_REP_NUM - 1; i > 0; i--)
rep[i] = rep[i-1];
rep[0] = sequence.offset;
/* Store the sequence */
ZSTD_storeSeq(seqStore, newLitLength, ip - newLitLength, iend,
OFFSET_TO_OFFBASE(sequence.offset),
sequence.matchLength);
ip += sequence.matchLength;
}
}
/* Fill the tables for the block compressor */
ZSTD_ldm_limitTableUpdate(ms, ip);
ZSTD_ldm_fillFastTables(ms, ip);
/* Compress the last literals */
return blockCompressor(ms, seqStore, rep, ip, iend - ip);
}

117
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_ldm.h vendored Normal file
View file

@ -0,0 +1,117 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_LDM_H
#define ZSTD_LDM_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "zstd_compress_internal.h" /* ldmParams_t, U32 */
#include "../zstd.h" /* ZSTD_CCtx, size_t */
/*-*************************************
* Long distance matching
***************************************/
#define ZSTD_LDM_DEFAULT_WINDOW_LOG ZSTD_WINDOWLOG_LIMIT_DEFAULT
void ZSTD_ldm_fillHashTable(
ldmState_t* state, const BYTE* ip,
const BYTE* iend, ldmParams_t const* params);
/**
* ZSTD_ldm_generateSequences():
*
* Generates the sequences using the long distance match finder.
* Generates long range matching sequences in `sequences`, which parse a prefix
* of the source. `sequences` must be large enough to store every sequence,
* which can be checked with `ZSTD_ldm_getMaxNbSeq()`.
* @returns 0 or an error code.
*
* NOTE: The user must have called ZSTD_window_update() for all of the input
* they have, even if they pass it to ZSTD_ldm_generateSequences() in chunks.
* NOTE: This function returns an error if it runs out of space to store
* sequences.
*/
size_t ZSTD_ldm_generateSequences(
ldmState_t* ldms, rawSeqStore_t* sequences,
ldmParams_t const* params, void const* src, size_t srcSize);
/**
* ZSTD_ldm_blockCompress():
*
* Compresses a block using the predefined sequences, along with a secondary
* block compressor. The literals section of every sequence is passed to the
* secondary block compressor, and those sequences are interspersed with the
* predefined sequences. Returns the length of the last literals.
* Updates `rawSeqStore.pos` to indicate how many sequences have been consumed.
* `rawSeqStore.seq` may also be updated to split the last sequence between two
* blocks.
* @return The length of the last literals.
*
* NOTE: The source must be at most the maximum block size, but the predefined
* sequences can be any size, and may be longer than the block. In the case that
* they are longer than the block, the last sequences may need to be split into
* two. We handle that case correctly, and update `rawSeqStore` appropriately.
* NOTE: This function does not return any errors.
*/
size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_paramSwitch_e useRowMatchFinder,
void const* src, size_t srcSize);
/**
* ZSTD_ldm_skipSequences():
*
* Skip past `srcSize` bytes worth of sequences in `rawSeqStore`.
* Avoids emitting matches less than `minMatch` bytes.
* Must be called for data that is not passed to ZSTD_ldm_blockCompress().
*/
void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize,
U32 const minMatch);
/* ZSTD_ldm_skipRawSeqStoreBytes():
* Moves forward in rawSeqStore by nbBytes, updating fields 'pos' and 'posInSequence'.
* Not to be used in conjunction with ZSTD_ldm_skipSequences().
* Must be called for data with is not passed to ZSTD_ldm_blockCompress().
*/
void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes);
/** ZSTD_ldm_getTableSize() :
* Estimate the space needed for long distance matching tables or 0 if LDM is
* disabled.
*/
size_t ZSTD_ldm_getTableSize(ldmParams_t params);
/** ZSTD_ldm_getSeqSpace() :
* Return an upper bound on the number of sequences that can be produced by
* the long distance matcher, or 0 if LDM is disabled.
*/
size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize);
/** ZSTD_ldm_adjustParameters() :
* If the params->hashRateLog is not set, set it to its default value based on
* windowLog and params->hashLog.
*
* Ensures that params->bucketSizeLog is <= params->hashLog (setting it to
* params->hashLog if it is not).
*
* Ensures that the minMatchLength >= targetLength during optimal parsing.
*/
void ZSTD_ldm_adjustParameters(ldmParams_t* params,
ZSTD_compressionParameters const* cParams);
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_FAST_H */

106
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_ldm_geartab.h vendored Normal file
View file

@ -0,0 +1,106 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_LDM_GEARTAB_H
#define ZSTD_LDM_GEARTAB_H
#include "../common/compiler.h" /* UNUSED_ATTR */
#include "../common/mem.h" /* U64 */
static UNUSED_ATTR const U64 ZSTD_ldm_gearTab[256] = {
0xf5b8f72c5f77775c, 0x84935f266b7ac412, 0xb647ada9ca730ccc,
0xb065bb4b114fb1de, 0x34584e7e8c3a9fd0, 0x4e97e17c6ae26b05,
0x3a03d743bc99a604, 0xcecd042422c4044f, 0x76de76c58524259e,
0x9c8528f65badeaca, 0x86563706e2097529, 0x2902475fa375d889,
0xafb32a9739a5ebe6, 0xce2714da3883e639, 0x21eaf821722e69e,
0x37b628620b628, 0x49a8d455d88caf5, 0x8556d711e6958140,
0x4f7ae74fc605c1f, 0x829f0c3468bd3a20, 0x4ffdc885c625179e,
0x8473de048a3daf1b, 0x51008822b05646b2, 0x69d75d12b2d1cc5f,
0x8c9d4a19159154bc, 0xc3cc10f4abbd4003, 0xd06ddc1cecb97391,
0xbe48e6e7ed80302e, 0x3481db31cee03547, 0xacc3f67cdaa1d210,
0x65cb771d8c7f96cc, 0x8eb27177055723dd, 0xc789950d44cd94be,
0x934feadc3700b12b, 0x5e485f11edbdf182, 0x1e2e2a46fd64767a,
0x2969ca71d82efa7c, 0x9d46e9935ebbba2e, 0xe056b67e05e6822b,
0x94d73f55739d03a0, 0xcd7010bdb69b5a03, 0x455ef9fcd79b82f4,
0x869cb54a8749c161, 0x38d1a4fa6185d225, 0xb475166f94bbe9bb,
0xa4143548720959f1, 0x7aed4780ba6b26ba, 0xd0ce264439e02312,
0x84366d746078d508, 0xa8ce973c72ed17be, 0x21c323a29a430b01,
0x9962d617e3af80ee, 0xab0ce91d9c8cf75b, 0x530e8ee6d19a4dbc,
0x2ef68c0cf53f5d72, 0xc03a681640a85506, 0x496e4e9f9c310967,
0x78580472b59b14a0, 0x273824c23b388577, 0x66bf923ad45cb553,
0x47ae1a5a2492ba86, 0x35e304569e229659, 0x4765182a46870b6f,
0x6cbab625e9099412, 0xddac9a2e598522c1, 0x7172086e666624f2,
0xdf5003ca503b7837, 0x88c0c1db78563d09, 0x58d51865acfc289d,
0x177671aec65224f1, 0xfb79d8a241e967d7, 0x2be1e101cad9a49a,
0x6625682f6e29186b, 0x399553457ac06e50, 0x35dffb4c23abb74,
0x429db2591f54aade, 0xc52802a8037d1009, 0x6acb27381f0b25f3,
0xf45e2551ee4f823b, 0x8b0ea2d99580c2f7, 0x3bed519cbcb4e1e1,
0xff452823dbb010a, 0x9d42ed614f3dd267, 0x5b9313c06257c57b,
0xa114b8008b5e1442, 0xc1fe311c11c13d4b, 0x66e8763ea34c5568,
0x8b982af1c262f05d, 0xee8876faaa75fbb7, 0x8a62a4d0d172bb2a,
0xc13d94a3b7449a97, 0x6dbbba9dc15d037c, 0xc786101f1d92e0f1,
0xd78681a907a0b79b, 0xf61aaf2962c9abb9, 0x2cfd16fcd3cb7ad9,
0x868c5b6744624d21, 0x25e650899c74ddd7, 0xba042af4a7c37463,
0x4eb1a539465a3eca, 0xbe09dbf03b05d5ca, 0x774e5a362b5472ba,
0x47a1221229d183cd, 0x504b0ca18ef5a2df, 0xdffbdfbde2456eb9,
0x46cd2b2fbee34634, 0xf2aef8fe819d98c3, 0x357f5276d4599d61,
0x24a5483879c453e3, 0x88026889192b4b9, 0x28da96671782dbec,
0x4ef37c40588e9aaa, 0x8837b90651bc9fb3, 0xc164f741d3f0e5d6,
0xbc135a0a704b70ba, 0x69cd868f7622ada, 0xbc37ba89e0b9c0ab,
0x47c14a01323552f6, 0x4f00794bacee98bb, 0x7107de7d637a69d5,
0x88af793bb6f2255e, 0xf3c6466b8799b598, 0xc288c616aa7f3b59,
0x81ca63cf42fca3fd, 0x88d85ace36a2674b, 0xd056bd3792389e7,
0xe55c396c4e9dd32d, 0xbefb504571e6c0a6, 0x96ab32115e91e8cc,
0xbf8acb18de8f38d1, 0x66dae58801672606, 0x833b6017872317fb,
0xb87c16f2d1c92864, 0xdb766a74e58b669c, 0x89659f85c61417be,
0xc8daad856011ea0c, 0x76a4b565b6fe7eae, 0xa469d085f6237312,
0xaaf0365683a3e96c, 0x4dbb746f8424f7b8, 0x638755af4e4acc1,
0x3d7807f5bde64486, 0x17be6d8f5bbb7639, 0x903f0cd44dc35dc,
0x67b672eafdf1196c, 0xa676ff93ed4c82f1, 0x521d1004c5053d9d,
0x37ba9ad09ccc9202, 0x84e54d297aacfb51, 0xa0b4b776a143445,
0x820d471e20b348e, 0x1874383cb83d46dc, 0x97edeec7a1efe11c,
0xb330e50b1bdc42aa, 0x1dd91955ce70e032, 0xa514cdb88f2939d5,
0x2791233fd90db9d3, 0x7b670a4cc50f7a9b, 0x77c07d2a05c6dfa5,
0xe3778b6646d0a6fa, 0xb39c8eda47b56749, 0x933ed448addbef28,
0xaf846af6ab7d0bf4, 0xe5af208eb666e49, 0x5e6622f73534cd6a,
0x297daeca42ef5b6e, 0x862daef3d35539a6, 0xe68722498f8e1ea9,
0x981c53093dc0d572, 0xfa09b0bfbf86fbf5, 0x30b1e96166219f15,
0x70e7d466bdc4fb83, 0x5a66736e35f2a8e9, 0xcddb59d2b7c1baef,
0xd6c7d247d26d8996, 0xea4e39eac8de1ba3, 0x539c8bb19fa3aff2,
0x9f90e4c5fd508d8, 0xa34e5956fbaf3385, 0x2e2f8e151d3ef375,
0x173691e9b83faec1, 0xb85a8d56bf016379, 0x8382381267408ae3,
0xb90f901bbdc0096d, 0x7c6ad32933bcec65, 0x76bb5e2f2c8ad595,
0x390f851a6cf46d28, 0xc3e6064da1c2da72, 0xc52a0c101cfa5389,
0xd78eaf84a3fbc530, 0x3781b9e2288b997e, 0x73c2f6dea83d05c4,
0x4228e364c5b5ed7, 0x9d7a3edf0da43911, 0x8edcfeda24686756,
0x5e7667a7b7a9b3a1, 0x4c4f389fa143791d, 0xb08bc1023da7cddc,
0x7ab4be3ae529b1cc, 0x754e6132dbe74ff9, 0x71635442a839df45,
0x2f6fb1643fbe52de, 0x961e0a42cf7a8177, 0xf3b45d83d89ef2ea,
0xee3de4cf4a6e3e9b, 0xcd6848542c3295e7, 0xe4cee1664c78662f,
0x9947548b474c68c4, 0x25d73777a5ed8b0b, 0xc915b1d636b7fc,
0x21c2ba75d9b0d2da, 0x5f6b5dcf608a64a1, 0xdcf333255ff9570c,
0x633b922418ced4ee, 0xc136dde0b004b34a, 0x58cc83b05d4b2f5a,
0x5eb424dda28e42d2, 0x62df47369739cd98, 0xb4e0b42485e4ce17,
0x16e1f0c1f9a8d1e7, 0x8ec3916707560ebf, 0x62ba6e2df2cc9db3,
0xcbf9f4ff77d83a16, 0x78d9d7d07d2bbcc4, 0xef554ce1e02c41f4,
0x8d7581127eccf94d, 0xa9b53336cb3c8a05, 0x38c42c0bf45c4f91,
0x640893cdf4488863, 0x80ec34bc575ea568, 0x39f324f5b48eaa40,
0xe9d9ed1f8eff527f, 0x9224fc058cc5a214, 0xbaba00b04cfe7741,
0x309a9f120fcf52af, 0xa558f3ec65626212, 0x424bec8b7adabe2f,
0x41622513a6aea433, 0xb88da2d5324ca798, 0xd287733b245528a4,
0x9a44697e6d68aec3, 0x7b1093be2f49bb28, 0x50bbec632e3d8aad,
0x6cd90723e1ea8283, 0x897b9e7431b02bf3, 0x219efdcb338a7047,
0x3b0311f0a27c0656, 0xdb17bf91c0db96e7, 0x8cd4fd6b4e85a5b2,
0xfab071054ba6409d, 0x40d6fe831fa9dfd9, 0xaf358debad7d791e,
0xeb8d0e25a65e3e58, 0xbbcbd3df14e08580, 0xcf751f27ecdab2b,
0x2b4da14f2613d8f4
};
#endif /* ZSTD_LDM_GEARTAB_H */

1472
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_opt.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

56
raytracer/nvpro_core/third_party/zstd/lib/compress/zstd_opt.h vendored Normal file
View file

@ -0,0 +1,56 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_OPT_H
#define ZSTD_OPT_H
#if defined (__cplusplus)
extern "C" {
#endif
#include "zstd_compress_internal.h"
/* used in ZSTD_loadDictionaryContent() */
void ZSTD_updateTree(ZSTD_matchState_t* ms, const BYTE* ip, const BYTE* iend);
size_t ZSTD_compressBlock_btopt(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btultra(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btultra2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btopt_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btultra_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btopt_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btultra_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
/* note : no btultra2 variant for extDict nor dictMatchState,
* because btultra2 is not meant to work with dictionaries
* and is only specific for the first block (no prefix) */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_OPT_H */

1867
raytracer/nvpro_core/third_party/zstd/lib/compress/zstdmt_compress.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

113
raytracer/nvpro_core/third_party/zstd/lib/compress/zstdmt_compress.h vendored Normal file
View file

@ -0,0 +1,113 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTDMT_COMPRESS_H
#define ZSTDMT_COMPRESS_H
#if defined (__cplusplus)
extern "C" {
#endif
/* Note : This is an internal API.
* These APIs used to be exposed with ZSTDLIB_API,
* because it used to be the only way to invoke MT compression.
* Now, you must use ZSTD_compress2 and ZSTD_compressStream2() instead.
*
* This API requires ZSTD_MULTITHREAD to be defined during compilation,
* otherwise ZSTDMT_createCCtx*() will fail.
*/
/* === Dependencies === */
#include "../common/zstd_deps.h" /* size_t */
#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_parameters */
#include "../zstd.h" /* ZSTD_inBuffer, ZSTD_outBuffer, ZSTDLIB_API */
/* === Constants === */
#ifndef ZSTDMT_NBWORKERS_MAX /* a different value can be selected at compile time */
# define ZSTDMT_NBWORKERS_MAX ((sizeof(void*)==4) /*32-bit*/ ? 64 : 256)
#endif
#ifndef ZSTDMT_JOBSIZE_MIN /* a different value can be selected at compile time */
# define ZSTDMT_JOBSIZE_MIN (512 KB)
#endif
#define ZSTDMT_JOBLOG_MAX (MEM_32bits() ? 29 : 30)
#define ZSTDMT_JOBSIZE_MAX (MEM_32bits() ? (512 MB) : (1024 MB))
/* ========================================================
* === Private interface, for use by ZSTD_compress.c ===
* === Not exposed in libzstd. Never invoke directly ===
* ======================================================== */
/* === Memory management === */
typedef struct ZSTDMT_CCtx_s ZSTDMT_CCtx;
/* Requires ZSTD_MULTITHREAD to be defined during compilation, otherwise it will return NULL. */
ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers,
ZSTD_customMem cMem,
ZSTD_threadPool *pool);
size_t ZSTDMT_freeCCtx(ZSTDMT_CCtx* mtctx);
size_t ZSTDMT_sizeof_CCtx(ZSTDMT_CCtx* mtctx);
/* === Streaming functions === */
size_t ZSTDMT_nextInputSizeHint(const ZSTDMT_CCtx* mtctx);
/*! ZSTDMT_initCStream_internal() :
* Private use only. Init streaming operation.
* expects params to be valid.
* must receive dict, or cdict, or none, but not both.
* mtctx can be freshly constructed or reused from a prior compression.
* If mtctx is reused, memory allocations from the prior compression may not be freed,
* even if they are not needed for the current compression.
* @return : 0, or an error code */
size_t ZSTDMT_initCStream_internal(ZSTDMT_CCtx* mtctx,
const void* dict, size_t dictSize, ZSTD_dictContentType_e dictContentType,
const ZSTD_CDict* cdict,
ZSTD_CCtx_params params, unsigned long long pledgedSrcSize);
/*! ZSTDMT_compressStream_generic() :
* Combines ZSTDMT_compressStream() with optional ZSTDMT_flushStream() or ZSTDMT_endStream()
* depending on flush directive.
* @return : minimum amount of data still to be flushed
* 0 if fully flushed
* or an error code
* note : needs to be init using any ZSTD_initCStream*() variant */
size_t ZSTDMT_compressStream_generic(ZSTDMT_CCtx* mtctx,
ZSTD_outBuffer* output,
ZSTD_inBuffer* input,
ZSTD_EndDirective endOp);
/*! ZSTDMT_toFlushNow()
* Tell how many bytes are ready to be flushed immediately.
* Probe the oldest active job (not yet entirely flushed) and check its output buffer.
* If return 0, it means there is no active job,
* or, it means oldest job is still active, but everything produced has been flushed so far,
* therefore flushing is limited by speed of oldest job. */
size_t ZSTDMT_toFlushNow(ZSTDMT_CCtx* mtctx);
/*! ZSTDMT_updateCParams_whileCompressing() :
* Updates only a selected set of compression parameters, to remain compatible with current frame.
* New parameters will be applied to next compression job. */
void ZSTDMT_updateCParams_whileCompressing(ZSTDMT_CCtx* mtctx, const ZSTD_CCtx_params* cctxParams);
/*! ZSTDMT_getFrameProgression():
* tells how much data has been consumed (input) and produced (output) for current frame.
* able to count progression inside worker threads.
*/
ZSTD_frameProgression ZSTDMT_getFrameProgression(ZSTDMT_CCtx* mtctx);
#if defined (__cplusplus)
}
#endif
#endif /* ZSTDMT_COMPRESS_H */

1882
raytracer/nvpro_core/third_party/zstd/lib/decompress/huf_decompress.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

576
raytracer/nvpro_core/third_party/zstd/lib/decompress/huf_decompress_amd64.S vendored Normal file
View file

@ -0,0 +1,576 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#include "../common/portability_macros.h"
/* Stack marking
* ref: https://wiki.gentoo.org/wiki/Hardened/GNU_stack_quickstart
*/
#if defined(__ELF__) && defined(__GNUC__)
.section .note.GNU-stack,"",%progbits
#endif
#if ZSTD_ENABLE_ASM_X86_64_BMI2
/* Calling convention:
*
* %rdi contains the first argument: HUF_DecompressAsmArgs*.
* %rbp isn't maintained (no frame pointer).
* %rsp contains the stack pointer that grows down.
* No red-zone is assumed, only addresses >= %rsp are used.
* All register contents are preserved.
*
* TODO: Support Windows calling convention.
*/
ZSTD_HIDE_ASM_FUNCTION(HUF_decompress4X1_usingDTable_internal_fast_asm_loop)
ZSTD_HIDE_ASM_FUNCTION(HUF_decompress4X2_usingDTable_internal_fast_asm_loop)
ZSTD_HIDE_ASM_FUNCTION(_HUF_decompress4X2_usingDTable_internal_fast_asm_loop)
ZSTD_HIDE_ASM_FUNCTION(_HUF_decompress4X1_usingDTable_internal_fast_asm_loop)
.global HUF_decompress4X1_usingDTable_internal_fast_asm_loop
.global HUF_decompress4X2_usingDTable_internal_fast_asm_loop
.global _HUF_decompress4X1_usingDTable_internal_fast_asm_loop
.global _HUF_decompress4X2_usingDTable_internal_fast_asm_loop
.text
/* Sets up register mappings for clarity.
* op[], bits[], dtable & ip[0] each get their own register.
* ip[1,2,3] & olimit alias var[].
* %rax is a scratch register.
*/
#define op0 rsi
#define op1 rbx
#define op2 rcx
#define op3 rdi
#define ip0 r8
#define ip1 r9
#define ip2 r10
#define ip3 r11
#define bits0 rbp
#define bits1 rdx
#define bits2 r12
#define bits3 r13
#define dtable r14
#define olimit r15
/* var[] aliases ip[1,2,3] & olimit
* ip[1,2,3] are saved every iteration.
* olimit is only used in compute_olimit.
*/
#define var0 r15
#define var1 r9
#define var2 r10
#define var3 r11
/* 32-bit var registers */
#define vard0 r15d
#define vard1 r9d
#define vard2 r10d
#define vard3 r11d
/* Calls X(N) for each stream 0, 1, 2, 3. */
#define FOR_EACH_STREAM(X) \
X(0); \
X(1); \
X(2); \
X(3)
/* Calls X(N, idx) for each stream 0, 1, 2, 3. */
#define FOR_EACH_STREAM_WITH_INDEX(X, idx) \
X(0, idx); \
X(1, idx); \
X(2, idx); \
X(3, idx)
/* Define both _HUF_* & HUF_* symbols because MacOS
* C symbols are prefixed with '_' & Linux symbols aren't.
*/
_HUF_decompress4X1_usingDTable_internal_fast_asm_loop:
HUF_decompress4X1_usingDTable_internal_fast_asm_loop:
ZSTD_CET_ENDBRANCH
/* Save all registers - even if they are callee saved for simplicity. */
push %rax
push %rbx
push %rcx
push %rdx
push %rbp
push %rsi
push %rdi
push %r8
push %r9
push %r10
push %r11
push %r12
push %r13
push %r14
push %r15
/* Read HUF_DecompressAsmArgs* args from %rax */
movq %rdi, %rax
movq 0(%rax), %ip0
movq 8(%rax), %ip1
movq 16(%rax), %ip2
movq 24(%rax), %ip3
movq 32(%rax), %op0
movq 40(%rax), %op1
movq 48(%rax), %op2
movq 56(%rax), %op3
movq 64(%rax), %bits0
movq 72(%rax), %bits1
movq 80(%rax), %bits2
movq 88(%rax), %bits3
movq 96(%rax), %dtable
push %rax /* argument */
push 104(%rax) /* ilimit */
push 112(%rax) /* oend */
push %olimit /* olimit space */
subq $24, %rsp
.L_4X1_compute_olimit:
/* Computes how many iterations we can do safely
* %r15, %rax may be clobbered
* rbx, rdx must be saved
* op3 & ip0 mustn't be clobbered
*/
movq %rbx, 0(%rsp)
movq %rdx, 8(%rsp)
movq 32(%rsp), %rax /* rax = oend */
subq %op3, %rax /* rax = oend - op3 */
/* r15 = (oend - op3) / 5 */
movabsq $-3689348814741910323, %rdx
mulq %rdx
movq %rdx, %r15
shrq $2, %r15
movq %ip0, %rax /* rax = ip0 */
movq 40(%rsp), %rdx /* rdx = ilimit */
subq %rdx, %rax /* rax = ip0 - ilimit */
movq %rax, %rbx /* rbx = ip0 - ilimit */
/* rdx = (ip0 - ilimit) / 7 */
movabsq $2635249153387078803, %rdx
mulq %rdx
subq %rdx, %rbx
shrq %rbx
addq %rbx, %rdx
shrq $2, %rdx
/* r15 = min(%rdx, %r15) */
cmpq %rdx, %r15
cmova %rdx, %r15
/* r15 = r15 * 5 */
leaq (%r15, %r15, 4), %r15
/* olimit = op3 + r15 */
addq %op3, %olimit
movq 8(%rsp), %rdx
movq 0(%rsp), %rbx
/* If (op3 + 20 > olimit) */
movq %op3, %rax /* rax = op3 */
addq $20, %rax /* rax = op3 + 20 */
cmpq %rax, %olimit /* op3 + 20 > olimit */
jb .L_4X1_exit
/* If (ip1 < ip0) go to exit */
cmpq %ip0, %ip1
jb .L_4X1_exit
/* If (ip2 < ip1) go to exit */
cmpq %ip1, %ip2
jb .L_4X1_exit
/* If (ip3 < ip2) go to exit */
cmpq %ip2, %ip3
jb .L_4X1_exit
/* Reads top 11 bits from bits[n]
* Loads dt[bits[n]] into var[n]
*/
#define GET_NEXT_DELT(n) \
movq $53, %var##n; \
shrxq %var##n, %bits##n, %var##n; \
movzwl (%dtable,%var##n,2),%vard##n
/* var[n] must contain the DTable entry computed with GET_NEXT_DELT
* Moves var[n] to %rax
* bits[n] <<= var[n] & 63
* op[n][idx] = %rax >> 8
* %ah is a way to access bits [8, 16) of %rax
*/
#define DECODE_FROM_DELT(n, idx) \
movq %var##n, %rax; \
shlxq %var##n, %bits##n, %bits##n; \
movb %ah, idx(%op##n)
/* Assumes GET_NEXT_DELT has been called.
* Calls DECODE_FROM_DELT then GET_NEXT_DELT
*/
#define DECODE_AND_GET_NEXT(n, idx) \
DECODE_FROM_DELT(n, idx); \
GET_NEXT_DELT(n) \
/* // ctz & nbBytes is stored in bits[n]
* // nbBits is stored in %rax
* ctz = CTZ[bits[n]]
* nbBits = ctz & 7
* nbBytes = ctz >> 3
* op[n] += 5
* ip[n] -= nbBytes
* // Note: x86-64 is little-endian ==> no bswap
* bits[n] = MEM_readST(ip[n]) | 1
* bits[n] <<= nbBits
*/
#define RELOAD_BITS(n) \
bsfq %bits##n, %bits##n; \
movq %bits##n, %rax; \
andq $7, %rax; \
shrq $3, %bits##n; \
leaq 5(%op##n), %op##n; \
subq %bits##n, %ip##n; \
movq (%ip##n), %bits##n; \
orq $1, %bits##n; \
shlx %rax, %bits##n, %bits##n
/* Store clobbered variables on the stack */
movq %olimit, 24(%rsp)
movq %ip1, 0(%rsp)
movq %ip2, 8(%rsp)
movq %ip3, 16(%rsp)
/* Call GET_NEXT_DELT for each stream */
FOR_EACH_STREAM(GET_NEXT_DELT)
.p2align 6
.L_4X1_loop_body:
/* Decode 5 symbols in each of the 4 streams (20 total)
* Must have called GET_NEXT_DELT for each stream
*/
FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 0)
FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 1)
FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 2)
FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 3)
FOR_EACH_STREAM_WITH_INDEX(DECODE_FROM_DELT, 4)
/* Load ip[1,2,3] from stack (var[] aliases them)
* ip[] is needed for RELOAD_BITS
* Each will be stored back to the stack after RELOAD
*/
movq 0(%rsp), %ip1
movq 8(%rsp), %ip2
movq 16(%rsp), %ip3
/* Reload each stream & fetch the next table entry
* to prepare for the next iteration
*/
RELOAD_BITS(0)
GET_NEXT_DELT(0)
RELOAD_BITS(1)
movq %ip1, 0(%rsp)
GET_NEXT_DELT(1)
RELOAD_BITS(2)
movq %ip2, 8(%rsp)
GET_NEXT_DELT(2)
RELOAD_BITS(3)
movq %ip3, 16(%rsp)
GET_NEXT_DELT(3)
/* If op3 < olimit: continue the loop */
cmp %op3, 24(%rsp)
ja .L_4X1_loop_body
/* Reload ip[1,2,3] from stack */
movq 0(%rsp), %ip1
movq 8(%rsp), %ip2
movq 16(%rsp), %ip3
/* Re-compute olimit */
jmp .L_4X1_compute_olimit
#undef GET_NEXT_DELT
#undef DECODE_FROM_DELT
#undef DECODE
#undef RELOAD_BITS
.L_4X1_exit:
addq $24, %rsp
/* Restore stack (oend & olimit) */
pop %rax /* olimit */
pop %rax /* oend */
pop %rax /* ilimit */
pop %rax /* arg */
/* Save ip / op / bits */
movq %ip0, 0(%rax)
movq %ip1, 8(%rax)
movq %ip2, 16(%rax)
movq %ip3, 24(%rax)
movq %op0, 32(%rax)
movq %op1, 40(%rax)
movq %op2, 48(%rax)
movq %op3, 56(%rax)
movq %bits0, 64(%rax)
movq %bits1, 72(%rax)
movq %bits2, 80(%rax)
movq %bits3, 88(%rax)
/* Restore registers */
pop %r15
pop %r14
pop %r13
pop %r12
pop %r11
pop %r10
pop %r9
pop %r8
pop %rdi
pop %rsi
pop %rbp
pop %rdx
pop %rcx
pop %rbx
pop %rax
ret
_HUF_decompress4X2_usingDTable_internal_fast_asm_loop:
HUF_decompress4X2_usingDTable_internal_fast_asm_loop:
ZSTD_CET_ENDBRANCH
/* Save all registers - even if they are callee saved for simplicity. */
push %rax
push %rbx
push %rcx
push %rdx
push %rbp
push %rsi
push %rdi
push %r8
push %r9
push %r10
push %r11
push %r12
push %r13
push %r14
push %r15
movq %rdi, %rax
movq 0(%rax), %ip0
movq 8(%rax), %ip1
movq 16(%rax), %ip2
movq 24(%rax), %ip3
movq 32(%rax), %op0
movq 40(%rax), %op1
movq 48(%rax), %op2
movq 56(%rax), %op3
movq 64(%rax), %bits0
movq 72(%rax), %bits1
movq 80(%rax), %bits2
movq 88(%rax), %bits3
movq 96(%rax), %dtable
push %rax /* argument */
push %rax /* olimit */
push 104(%rax) /* ilimit */
movq 112(%rax), %rax
push %rax /* oend3 */
movq %op3, %rax
push %rax /* oend2 */
movq %op2, %rax
push %rax /* oend1 */
movq %op1, %rax
push %rax /* oend0 */
/* Scratch space */
subq $8, %rsp
.L_4X2_compute_olimit:
/* Computes how many iterations we can do safely
* %r15, %rax may be clobbered
* rdx must be saved
* op[1,2,3,4] & ip0 mustn't be clobbered
*/
movq %rdx, 0(%rsp)
/* We can consume up to 7 input bytes each iteration. */
movq %ip0, %rax /* rax = ip0 */
movq 40(%rsp), %rdx /* rdx = ilimit */
subq %rdx, %rax /* rax = ip0 - ilimit */
movq %rax, %r15 /* r15 = ip0 - ilimit */
/* rdx = rax / 7 */
movabsq $2635249153387078803, %rdx
mulq %rdx
subq %rdx, %r15
shrq %r15
addq %r15, %rdx
shrq $2, %rdx
/* r15 = (ip0 - ilimit) / 7 */
movq %rdx, %r15
/* r15 = min(r15, min(oend0 - op0, oend1 - op1, oend2 - op2, oend3 - op3) / 10) */
movq 8(%rsp), %rax /* rax = oend0 */
subq %op0, %rax /* rax = oend0 - op0 */
movq 16(%rsp), %rdx /* rdx = oend1 */
subq %op1, %rdx /* rdx = oend1 - op1 */
cmpq %rax, %rdx
cmova %rax, %rdx /* rdx = min(%rdx, %rax) */
movq 24(%rsp), %rax /* rax = oend2 */
subq %op2, %rax /* rax = oend2 - op2 */
cmpq %rax, %rdx
cmova %rax, %rdx /* rdx = min(%rdx, %rax) */
movq 32(%rsp), %rax /* rax = oend3 */
subq %op3, %rax /* rax = oend3 - op3 */
cmpq %rax, %rdx
cmova %rax, %rdx /* rdx = min(%rdx, %rax) */
movabsq $-3689348814741910323, %rax
mulq %rdx
shrq $3, %rdx /* rdx = rdx / 10 */
/* r15 = min(%rdx, %r15) */
cmpq %rdx, %r15
cmova %rdx, %r15
/* olimit = op3 + 5 * r15 */
movq %r15, %rax
leaq (%op3, %rax, 4), %olimit
addq %rax, %olimit
movq 0(%rsp), %rdx
/* If (op3 + 10 > olimit) */
movq %op3, %rax /* rax = op3 */
addq $10, %rax /* rax = op3 + 10 */
cmpq %rax, %olimit /* op3 + 10 > olimit */
jb .L_4X2_exit
/* If (ip1 < ip0) go to exit */
cmpq %ip0, %ip1
jb .L_4X2_exit
/* If (ip2 < ip1) go to exit */
cmpq %ip1, %ip2
jb .L_4X2_exit
/* If (ip3 < ip2) go to exit */
cmpq %ip2, %ip3
jb .L_4X2_exit
#define DECODE(n, idx) \
movq %bits##n, %rax; \
shrq $53, %rax; \
movzwl 0(%dtable,%rax,4),%r8d; \
movzbl 2(%dtable,%rax,4),%r15d; \
movzbl 3(%dtable,%rax,4),%eax; \
movw %r8w, (%op##n); \
shlxq %r15, %bits##n, %bits##n; \
addq %rax, %op##n
#define RELOAD_BITS(n) \
bsfq %bits##n, %bits##n; \
movq %bits##n, %rax; \
shrq $3, %bits##n; \
andq $7, %rax; \
subq %bits##n, %ip##n; \
movq (%ip##n), %bits##n; \
orq $1, %bits##n; \
shlxq %rax, %bits##n, %bits##n
movq %olimit, 48(%rsp)
.p2align 6
.L_4X2_loop_body:
/* We clobber r8, so store it on the stack */
movq %r8, 0(%rsp)
/* Decode 5 symbols from each of the 4 streams (20 symbols total). */
FOR_EACH_STREAM_WITH_INDEX(DECODE, 0)
FOR_EACH_STREAM_WITH_INDEX(DECODE, 1)
FOR_EACH_STREAM_WITH_INDEX(DECODE, 2)
FOR_EACH_STREAM_WITH_INDEX(DECODE, 3)
FOR_EACH_STREAM_WITH_INDEX(DECODE, 4)
/* Reload r8 */
movq 0(%rsp), %r8
FOR_EACH_STREAM(RELOAD_BITS)
cmp %op3, 48(%rsp)
ja .L_4X2_loop_body
jmp .L_4X2_compute_olimit
#undef DECODE
#undef RELOAD_BITS
.L_4X2_exit:
addq $8, %rsp
/* Restore stack (oend & olimit) */
pop %rax /* oend0 */
pop %rax /* oend1 */
pop %rax /* oend2 */
pop %rax /* oend3 */
pop %rax /* ilimit */
pop %rax /* olimit */
pop %rax /* arg */
/* Save ip / op / bits */
movq %ip0, 0(%rax)
movq %ip1, 8(%rax)
movq %ip2, 16(%rax)
movq %ip3, 24(%rax)
movq %op0, 32(%rax)
movq %op1, 40(%rax)
movq %op2, 48(%rax)
movq %op3, 56(%rax)
movq %bits0, 64(%rax)
movq %bits1, 72(%rax)
movq %bits2, 80(%rax)
movq %bits3, 88(%rax)
/* Restore registers */
pop %r15
pop %r14
pop %r13
pop %r12
pop %r11
pop %r10
pop %r9
pop %r8
pop %rdi
pop %rsi
pop %rbp
pop %rdx
pop %rcx
pop %rbx
pop %rax
ret
#endif

244
raytracer/nvpro_core/third_party/zstd/lib/decompress/zstd_ddict.c vendored Normal file
View file

@ -0,0 +1,244 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* zstd_ddict.c :
* concentrates all logic that needs to know the internals of ZSTD_DDict object */
/*-*******************************************************
* Dependencies
*********************************************************/
#include "../common/allocations.h" /* ZSTD_customMalloc, ZSTD_customFree */
#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */
#include "../common/cpu.h" /* bmi2 */
#include "../common/mem.h" /* low level memory routines */
#define FSE_STATIC_LINKING_ONLY
#include "../common/fse.h"
#include "../common/huf.h"
#include "zstd_decompress_internal.h"
#include "zstd_ddict.h"
#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1)
# include "../legacy/zstd_legacy.h"
#endif
/*-*******************************************************
* Types
*********************************************************/
struct ZSTD_DDict_s {
void* dictBuffer;
const void* dictContent;
size_t dictSize;
ZSTD_entropyDTables_t entropy;
U32 dictID;
U32 entropyPresent;
ZSTD_customMem cMem;
}; /* typedef'd to ZSTD_DDict within "zstd.h" */
const void* ZSTD_DDict_dictContent(const ZSTD_DDict* ddict)
{
assert(ddict != NULL);
return ddict->dictContent;
}
size_t ZSTD_DDict_dictSize(const ZSTD_DDict* ddict)
{
assert(ddict != NULL);
return ddict->dictSize;
}
void ZSTD_copyDDictParameters(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict)
{
DEBUGLOG(4, "ZSTD_copyDDictParameters");
assert(dctx != NULL);
assert(ddict != NULL);
dctx->dictID = ddict->dictID;
dctx->prefixStart = ddict->dictContent;
dctx->virtualStart = ddict->dictContent;
dctx->dictEnd = (const BYTE*)ddict->dictContent + ddict->dictSize;
dctx->previousDstEnd = dctx->dictEnd;
#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
dctx->dictContentBeginForFuzzing = dctx->prefixStart;
dctx->dictContentEndForFuzzing = dctx->previousDstEnd;
#endif
if (ddict->entropyPresent) {
dctx->litEntropy = 1;
dctx->fseEntropy = 1;
dctx->LLTptr = ddict->entropy.LLTable;
dctx->MLTptr = ddict->entropy.MLTable;
dctx->OFTptr = ddict->entropy.OFTable;
dctx->HUFptr = ddict->entropy.hufTable;
dctx->entropy.rep[0] = ddict->entropy.rep[0];
dctx->entropy.rep[1] = ddict->entropy.rep[1];
dctx->entropy.rep[2] = ddict->entropy.rep[2];
} else {
dctx->litEntropy = 0;
dctx->fseEntropy = 0;
}
}
static size_t
ZSTD_loadEntropy_intoDDict(ZSTD_DDict* ddict,
ZSTD_dictContentType_e dictContentType)
{
ddict->dictID = 0;
ddict->entropyPresent = 0;
if (dictContentType == ZSTD_dct_rawContent) return 0;
if (ddict->dictSize < 8) {
if (dictContentType == ZSTD_dct_fullDict)
return ERROR(dictionary_corrupted); /* only accept specified dictionaries */
return 0; /* pure content mode */
}
{ U32 const magic = MEM_readLE32(ddict->dictContent);
if (magic != ZSTD_MAGIC_DICTIONARY) {
if (dictContentType == ZSTD_dct_fullDict)
return ERROR(dictionary_corrupted); /* only accept specified dictionaries */
return 0; /* pure content mode */
}
}
ddict->dictID = MEM_readLE32((const char*)ddict->dictContent + ZSTD_FRAMEIDSIZE);
/* load entropy tables */
RETURN_ERROR_IF(ZSTD_isError(ZSTD_loadDEntropy(
&ddict->entropy, ddict->dictContent, ddict->dictSize)),
dictionary_corrupted, "");
ddict->entropyPresent = 1;
return 0;
}
static size_t ZSTD_initDDict_internal(ZSTD_DDict* ddict,
const void* dict, size_t dictSize,
ZSTD_dictLoadMethod_e dictLoadMethod,
ZSTD_dictContentType_e dictContentType)
{
if ((dictLoadMethod == ZSTD_dlm_byRef) || (!dict) || (!dictSize)) {
ddict->dictBuffer = NULL;
ddict->dictContent = dict;
if (!dict) dictSize = 0;
} else {
void* const internalBuffer = ZSTD_customMalloc(dictSize, ddict->cMem);
ddict->dictBuffer = internalBuffer;
ddict->dictContent = internalBuffer;
if (!internalBuffer) return ERROR(memory_allocation);
ZSTD_memcpy(internalBuffer, dict, dictSize);
}
ddict->dictSize = dictSize;
ddict->entropy.hufTable[0] = (HUF_DTable)((ZSTD_HUFFDTABLE_CAPACITY_LOG)*0x1000001); /* cover both little and big endian */
/* parse dictionary content */
FORWARD_IF_ERROR( ZSTD_loadEntropy_intoDDict(ddict, dictContentType) , "");
return 0;
}
ZSTD_DDict* ZSTD_createDDict_advanced(const void* dict, size_t dictSize,
ZSTD_dictLoadMethod_e dictLoadMethod,
ZSTD_dictContentType_e dictContentType,
ZSTD_customMem customMem)
{
if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL;
{ ZSTD_DDict* const ddict = (ZSTD_DDict*) ZSTD_customMalloc(sizeof(ZSTD_DDict), customMem);
if (ddict == NULL) return NULL;
ddict->cMem = customMem;
{ size_t const initResult = ZSTD_initDDict_internal(ddict,
dict, dictSize,
dictLoadMethod, dictContentType);
if (ZSTD_isError(initResult)) {
ZSTD_freeDDict(ddict);
return NULL;
} }
return ddict;
}
}
/*! ZSTD_createDDict() :
* Create a digested dictionary, to start decompression without startup delay.
* `dict` content is copied inside DDict.
* Consequently, `dict` can be released after `ZSTD_DDict` creation */
ZSTD_DDict* ZSTD_createDDict(const void* dict, size_t dictSize)
{
ZSTD_customMem const allocator = { NULL, NULL, NULL };
return ZSTD_createDDict_advanced(dict, dictSize, ZSTD_dlm_byCopy, ZSTD_dct_auto, allocator);
}
/*! ZSTD_createDDict_byReference() :
* Create a digested dictionary, to start decompression without startup delay.
* Dictionary content is simply referenced, it will be accessed during decompression.
* Warning : dictBuffer must outlive DDict (DDict must be freed before dictBuffer) */
ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize)
{
ZSTD_customMem const allocator = { NULL, NULL, NULL };
return ZSTD_createDDict_advanced(dictBuffer, dictSize, ZSTD_dlm_byRef, ZSTD_dct_auto, allocator);
}
const ZSTD_DDict* ZSTD_initStaticDDict(
void* sBuffer, size_t sBufferSize,
const void* dict, size_t dictSize,
ZSTD_dictLoadMethod_e dictLoadMethod,
ZSTD_dictContentType_e dictContentType)
{
size_t const neededSpace = sizeof(ZSTD_DDict)
+ (dictLoadMethod == ZSTD_dlm_byRef ? 0 : dictSize);
ZSTD_DDict* const ddict = (ZSTD_DDict*)sBuffer;
assert(sBuffer != NULL);
assert(dict != NULL);
if ((size_t)sBuffer & 7) return NULL; /* 8-aligned */
if (sBufferSize < neededSpace) return NULL;
if (dictLoadMethod == ZSTD_dlm_byCopy) {
ZSTD_memcpy(ddict+1, dict, dictSize); /* local copy */
dict = ddict+1;
}
if (ZSTD_isError( ZSTD_initDDict_internal(ddict,
dict, dictSize,
ZSTD_dlm_byRef, dictContentType) ))
return NULL;
return ddict;
}
size_t ZSTD_freeDDict(ZSTD_DDict* ddict)
{
if (ddict==NULL) return 0; /* support free on NULL */
{ ZSTD_customMem const cMem = ddict->cMem;
ZSTD_customFree(ddict->dictBuffer, cMem);
ZSTD_customFree(ddict, cMem);
return 0;
}
}
/*! ZSTD_estimateDDictSize() :
* Estimate amount of memory that will be needed to create a dictionary for decompression.
* Note : dictionary created by reference using ZSTD_dlm_byRef are smaller */
size_t ZSTD_estimateDDictSize(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod)
{
return sizeof(ZSTD_DDict) + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : dictSize);
}
size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict)
{
if (ddict==NULL) return 0; /* support sizeof on NULL */
return sizeof(*ddict) + (ddict->dictBuffer ? ddict->dictSize : 0) ;
}
/*! ZSTD_getDictID_fromDDict() :
* Provides the dictID of the dictionary loaded into `ddict`.
* If @return == 0, the dictionary is not conformant to Zstandard specification, or empty.
* Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */
unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict)
{
if (ddict==NULL) return 0;
return ddict->dictID;
}

44
raytracer/nvpro_core/third_party/zstd/lib/decompress/zstd_ddict.h vendored Normal file
View file

@ -0,0 +1,44 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_DDICT_H
#define ZSTD_DDICT_H
/*-*******************************************************
* Dependencies
*********************************************************/
#include "../common/zstd_deps.h" /* size_t */
#include "../zstd.h" /* ZSTD_DDict, and several public functions */
/*-*******************************************************
* Interface
*********************************************************/
/* note: several prototypes are already published in `zstd.h` :
* ZSTD_createDDict()
* ZSTD_createDDict_byReference()
* ZSTD_createDDict_advanced()
* ZSTD_freeDDict()
* ZSTD_initStaticDDict()
* ZSTD_sizeof_DDict()
* ZSTD_estimateDDictSize()
* ZSTD_getDictID_fromDict()
*/
const void* ZSTD_DDict_dictContent(const ZSTD_DDict* ddict);
size_t ZSTD_DDict_dictSize(const ZSTD_DDict* ddict);
void ZSTD_copyDDictParameters(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict);
#endif /* ZSTD_DDICT_H */

2355
raytracer/nvpro_core/third_party/zstd/lib/decompress/zstd_decompress.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

2192
raytracer/nvpro_core/third_party/zstd/lib/decompress/zstd_decompress_block.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

73
raytracer/nvpro_core/third_party/zstd/lib/decompress/zstd_decompress_block.h vendored Normal file
View file

@ -0,0 +1,73 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_DEC_BLOCK_H
#define ZSTD_DEC_BLOCK_H
/*-*******************************************************
* Dependencies
*********************************************************/
#include "../common/zstd_deps.h" /* size_t */
#include "../zstd.h" /* DCtx, and some public functions */
#include "../common/zstd_internal.h" /* blockProperties_t, and some public functions */
#include "zstd_decompress_internal.h" /* ZSTD_seqSymbol */
/* === Prototypes === */
/* note: prototypes already published within `zstd.h` :
* ZSTD_decompressBlock()
*/
/* note: prototypes already published within `zstd_internal.h` :
* ZSTD_getcBlockSize()
* ZSTD_decodeSeqHeaders()
*/
/* Streaming state is used to inform allocation of the literal buffer */
typedef enum {
not_streaming = 0,
is_streaming = 1
} streaming_operation;
/* ZSTD_decompressBlock_internal() :
* decompress block, starting at `src`,
* into destination buffer `dst`.
* @return : decompressed block size,
* or an error code (which can be tested using ZSTD_isError())
*/
size_t ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize, const int frame, const streaming_operation streaming);
/* ZSTD_buildFSETable() :
* generate FSE decoding table for one symbol (ll, ml or off)
* this function must be called with valid parameters only
* (dt is large enough, normalizedCounter distribution total is a power of 2, max is within range, etc.)
* in which case it cannot fail.
* The workspace must be 4-byte aligned and at least ZSTD_BUILD_FSE_TABLE_WKSP_SIZE bytes, which is
* defined in zstd_decompress_internal.h.
* Internal use only.
*/
void ZSTD_buildFSETable(ZSTD_seqSymbol* dt,
const short* normalizedCounter, unsigned maxSymbolValue,
const U32* baseValue, const U8* nbAdditionalBits,
unsigned tableLog, void* wksp, size_t wkspSize,
int bmi2);
/* Internal definition of ZSTD_decompressBlock() to avoid deprecation warnings. */
size_t ZSTD_decompressBlock_deprecated(ZSTD_DCtx* dctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize);
#endif /* ZSTD_DEC_BLOCK_H */

238
raytracer/nvpro_core/third_party/zstd/lib/decompress/zstd_decompress_internal.h vendored Normal file
View file

@ -0,0 +1,238 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/* zstd_decompress_internal:
* objects and definitions shared within lib/decompress modules */
#ifndef ZSTD_DECOMPRESS_INTERNAL_H
#define ZSTD_DECOMPRESS_INTERNAL_H
/*-*******************************************************
* Dependencies
*********************************************************/
#include "../common/mem.h" /* BYTE, U16, U32 */
#include "../common/zstd_internal.h" /* constants : MaxLL, MaxML, MaxOff, LLFSELog, etc. */
/*-*******************************************************
* Constants
*********************************************************/
static UNUSED_ATTR const U32 LL_base[MaxLL+1] = {
0, 1, 2, 3, 4, 5, 6, 7,
8, 9, 10, 11, 12, 13, 14, 15,
16, 18, 20, 22, 24, 28, 32, 40,
48, 64, 0x80, 0x100, 0x200, 0x400, 0x800, 0x1000,
0x2000, 0x4000, 0x8000, 0x10000 };
static UNUSED_ATTR const U32 OF_base[MaxOff+1] = {
0, 1, 1, 5, 0xD, 0x1D, 0x3D, 0x7D,
0xFD, 0x1FD, 0x3FD, 0x7FD, 0xFFD, 0x1FFD, 0x3FFD, 0x7FFD,
0xFFFD, 0x1FFFD, 0x3FFFD, 0x7FFFD, 0xFFFFD, 0x1FFFFD, 0x3FFFFD, 0x7FFFFD,
0xFFFFFD, 0x1FFFFFD, 0x3FFFFFD, 0x7FFFFFD, 0xFFFFFFD, 0x1FFFFFFD, 0x3FFFFFFD, 0x7FFFFFFD };
static UNUSED_ATTR const U8 OF_bits[MaxOff+1] = {
0, 1, 2, 3, 4, 5, 6, 7,
8, 9, 10, 11, 12, 13, 14, 15,
16, 17, 18, 19, 20, 21, 22, 23,
24, 25, 26, 27, 28, 29, 30, 31 };
static UNUSED_ATTR const U32 ML_base[MaxML+1] = {
3, 4, 5, 6, 7, 8, 9, 10,
11, 12, 13, 14, 15, 16, 17, 18,
19, 20, 21, 22, 23, 24, 25, 26,
27, 28, 29, 30, 31, 32, 33, 34,
35, 37, 39, 41, 43, 47, 51, 59,
67, 83, 99, 0x83, 0x103, 0x203, 0x403, 0x803,
0x1003, 0x2003, 0x4003, 0x8003, 0x10003 };
/*-*******************************************************
* Decompression types
*********************************************************/
typedef struct {
U32 fastMode;
U32 tableLog;
} ZSTD_seqSymbol_header;
typedef struct {
U16 nextState;
BYTE nbAdditionalBits;
BYTE nbBits;
U32 baseValue;
} ZSTD_seqSymbol;
#define SEQSYMBOL_TABLE_SIZE(log) (1 + (1 << (log)))
#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE (sizeof(S16) * (MaxSeq + 1) + (1u << MaxFSELog) + sizeof(U64))
#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32 ((ZSTD_BUILD_FSE_TABLE_WKSP_SIZE + sizeof(U32) - 1) / sizeof(U32))
#define ZSTD_HUFFDTABLE_CAPACITY_LOG 12
typedef struct {
ZSTD_seqSymbol LLTable[SEQSYMBOL_TABLE_SIZE(LLFSELog)]; /* Note : Space reserved for FSE Tables */
ZSTD_seqSymbol OFTable[SEQSYMBOL_TABLE_SIZE(OffFSELog)]; /* is also used as temporary workspace while building hufTable during DDict creation */
ZSTD_seqSymbol MLTable[SEQSYMBOL_TABLE_SIZE(MLFSELog)]; /* and therefore must be at least HUF_DECOMPRESS_WORKSPACE_SIZE large */
HUF_DTable hufTable[HUF_DTABLE_SIZE(ZSTD_HUFFDTABLE_CAPACITY_LOG)]; /* can accommodate HUF_decompress4X */
U32 rep[ZSTD_REP_NUM];
U32 workspace[ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32];
} ZSTD_entropyDTables_t;
typedef enum { ZSTDds_getFrameHeaderSize, ZSTDds_decodeFrameHeader,
ZSTDds_decodeBlockHeader, ZSTDds_decompressBlock,
ZSTDds_decompressLastBlock, ZSTDds_checkChecksum,
ZSTDds_decodeSkippableHeader, ZSTDds_skipFrame } ZSTD_dStage;
typedef enum { zdss_init=0, zdss_loadHeader,
zdss_read, zdss_load, zdss_flush } ZSTD_dStreamStage;
typedef enum {
ZSTD_use_indefinitely = -1, /* Use the dictionary indefinitely */
ZSTD_dont_use = 0, /* Do not use the dictionary (if one exists free it) */
ZSTD_use_once = 1 /* Use the dictionary once and set to ZSTD_dont_use */
} ZSTD_dictUses_e;
/* Hashset for storing references to multiple ZSTD_DDict within ZSTD_DCtx */
typedef struct {
const ZSTD_DDict** ddictPtrTable;
size_t ddictPtrTableSize;
size_t ddictPtrCount;
} ZSTD_DDictHashSet;
#ifndef ZSTD_DECODER_INTERNAL_BUFFER
# define ZSTD_DECODER_INTERNAL_BUFFER (1 << 16)
#endif
#define ZSTD_LBMIN 64
#define ZSTD_LBMAX (128 << 10)
/* extra buffer, compensates when dst is not large enough to store litBuffer */
#define ZSTD_LITBUFFEREXTRASIZE BOUNDED(ZSTD_LBMIN, ZSTD_DECODER_INTERNAL_BUFFER, ZSTD_LBMAX)
typedef enum {
ZSTD_not_in_dst = 0, /* Stored entirely within litExtraBuffer */
ZSTD_in_dst = 1, /* Stored entirely within dst (in memory after current output write) */
ZSTD_split = 2 /* Split between litExtraBuffer and dst */
} ZSTD_litLocation_e;
struct ZSTD_DCtx_s
{
const ZSTD_seqSymbol* LLTptr;
const ZSTD_seqSymbol* MLTptr;
const ZSTD_seqSymbol* OFTptr;
const HUF_DTable* HUFptr;
ZSTD_entropyDTables_t entropy;
U32 workspace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; /* space needed when building huffman tables */
const void* previousDstEnd; /* detect continuity */
const void* prefixStart; /* start of current segment */
const void* virtualStart; /* virtual start of previous segment if it was just before current one */
const void* dictEnd; /* end of previous segment */
size_t expected;
ZSTD_frameHeader fParams;
U64 processedCSize;
U64 decodedSize;
blockType_e bType; /* used in ZSTD_decompressContinue(), store blockType between block header decoding and block decompression stages */
ZSTD_dStage stage;
U32 litEntropy;
U32 fseEntropy;
XXH64_state_t xxhState;
size_t headerSize;
ZSTD_format_e format;
ZSTD_forceIgnoreChecksum_e forceIgnoreChecksum; /* User specified: if == 1, will ignore checksums in compressed frame. Default == 0 */
U32 validateChecksum; /* if == 1, will validate checksum. Is == 1 if (fParams.checksumFlag == 1) and (forceIgnoreChecksum == 0). */
const BYTE* litPtr;
ZSTD_customMem customMem;
size_t litSize;
size_t rleSize;
size_t staticSize;
#if DYNAMIC_BMI2 != 0
int bmi2; /* == 1 if the CPU supports BMI2 and 0 otherwise. CPU support is determined dynamically once per context lifetime. */
#endif
/* dictionary */
ZSTD_DDict* ddictLocal;
const ZSTD_DDict* ddict; /* set by ZSTD_initDStream_usingDDict(), or ZSTD_DCtx_refDDict() */
U32 dictID;
int ddictIsCold; /* if == 1 : dictionary is "new" for working context, and presumed "cold" (not in cpu cache) */
ZSTD_dictUses_e dictUses;
ZSTD_DDictHashSet* ddictSet; /* Hash set for multiple ddicts */
ZSTD_refMultipleDDicts_e refMultipleDDicts; /* User specified: if == 1, will allow references to multiple DDicts. Default == 0 (disabled) */
int disableHufAsm;
/* streaming */
ZSTD_dStreamStage streamStage;
char* inBuff;
size_t inBuffSize;
size_t inPos;
size_t maxWindowSize;
char* outBuff;
size_t outBuffSize;
size_t outStart;
size_t outEnd;
size_t lhSize;
#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1)
void* legacyContext;
U32 previousLegacyVersion;
U32 legacyVersion;
#endif
U32 hostageByte;
int noForwardProgress;
ZSTD_bufferMode_e outBufferMode;
ZSTD_outBuffer expectedOutBuffer;
/* workspace */
BYTE* litBuffer;
const BYTE* litBufferEnd;
ZSTD_litLocation_e litBufferLocation;
BYTE litExtraBuffer[ZSTD_LITBUFFEREXTRASIZE + WILDCOPY_OVERLENGTH]; /* literal buffer can be split between storage within dst and within this scratch buffer */
BYTE headerBuffer[ZSTD_FRAMEHEADERSIZE_MAX];
size_t oversizedDuration;
#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION
void const* dictContentBeginForFuzzing;
void const* dictContentEndForFuzzing;
#endif
/* Tracing */
#if ZSTD_TRACE
ZSTD_TraceCtx traceCtx;
#endif
}; /* typedef'd to ZSTD_DCtx within "zstd.h" */
MEM_STATIC int ZSTD_DCtx_get_bmi2(const struct ZSTD_DCtx_s *dctx) {
#if DYNAMIC_BMI2 != 0
return dctx->bmi2;
#else
(void)dctx;
return 0;
#endif
}
/*-*******************************************************
* Shared internal functions
*********************************************************/
/*! ZSTD_loadDEntropy() :
* dict : must point at beginning of a valid zstd dictionary.
* @return : size of dictionary header (size of magic number + dict ID + entropy tables) */
size_t ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy,
const void* const dict, size_t const dictSize);
/*! ZSTD_checkContinuity() :
* check if next `dst` follows previous position, where decompression ended.
* If yes, do nothing (continue on current segment).
* If not, classify previous segment as "external dictionary", and start a new segment.
* This function cannot fail. */
void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst, size_t dstSize);
#endif /* ZSTD_DECOMPRESS_INTERNAL_H */

422
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_legacy.h vendored Normal file
View file

@ -0,0 +1,422 @@
/*
* Copyright (c) Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_LEGACY_H
#define ZSTD_LEGACY_H
#if defined (__cplusplus)
extern "C" {
#endif
/* *************************************
* Includes
***************************************/
#include "../common/mem.h" /* MEM_STATIC */
#include "../common/error_private.h" /* ERROR */
#include "../common/zstd_internal.h" /* ZSTD_inBuffer, ZSTD_outBuffer, ZSTD_frameSizeInfo */
#if !defined (ZSTD_LEGACY_SUPPORT) || (ZSTD_LEGACY_SUPPORT == 0)
# undef ZSTD_LEGACY_SUPPORT
# define ZSTD_LEGACY_SUPPORT 8
#endif
#if (ZSTD_LEGACY_SUPPORT <= 1)
# include "zstd_v01.h"
#endif
#if (ZSTD_LEGACY_SUPPORT <= 2)
# include "zstd_v02.h"
#endif
#if (ZSTD_LEGACY_SUPPORT <= 3)
# include "zstd_v03.h"
#endif
#if (ZSTD_LEGACY_SUPPORT <= 4)
# include "zstd_v04.h"
#endif
#if (ZSTD_LEGACY_SUPPORT <= 5)
# include "zstd_v05.h"
#endif
#if (ZSTD_LEGACY_SUPPORT <= 6)
# include "zstd_v06.h"
#endif
#if (ZSTD_LEGACY_SUPPORT <= 7)
# include "zstd_v07.h"
#endif
/** ZSTD_isLegacy() :
@return : > 0 if supported by legacy decoder. 0 otherwise.
return value is the version.
*/
MEM_STATIC unsigned ZSTD_isLegacy(const void* src, size_t srcSize)
{
U32 magicNumberLE;
if (srcSize<4) return 0;
magicNumberLE = MEM_readLE32(src);
switch(magicNumberLE)
{
#if (ZSTD_LEGACY_SUPPORT <= 1)
case ZSTDv01_magicNumberLE:return 1;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 2)
case ZSTDv02_magicNumber : return 2;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 3)
case ZSTDv03_magicNumber : return 3;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 4)
case ZSTDv04_magicNumber : return 4;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 5)
case ZSTDv05_MAGICNUMBER : return 5;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 6)
case ZSTDv06_MAGICNUMBER : return 6;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 7)
case ZSTDv07_MAGICNUMBER : return 7;
#endif
default : return 0;
}
}
MEM_STATIC unsigned long long ZSTD_getDecompressedSize_legacy(const void* src, size_t srcSize)
{
U32 const version = ZSTD_isLegacy(src, srcSize);
if (version < 5) return 0; /* no decompressed size in frame header, or not a legacy format */
#if (ZSTD_LEGACY_SUPPORT <= 5)
if (version==5) {
ZSTDv05_parameters fParams;
size_t const frResult = ZSTDv05_getFrameParams(&fParams, src, srcSize);
if (frResult != 0) return 0;
return fParams.srcSize;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 6)
if (version==6) {
ZSTDv06_frameParams fParams;
size_t const frResult = ZSTDv06_getFrameParams(&fParams, src, srcSize);
if (frResult != 0) return 0;
return fParams.frameContentSize;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 7)
if (version==7) {
ZSTDv07_frameParams fParams;
size_t const frResult = ZSTDv07_getFrameParams(&fParams, src, srcSize);
if (frResult != 0) return 0;
return fParams.frameContentSize;
}
#endif
return 0; /* should not be possible */
}
MEM_STATIC size_t ZSTD_decompressLegacy(
void* dst, size_t dstCapacity,
const void* src, size_t compressedSize,
const void* dict,size_t dictSize)
{
U32 const version = ZSTD_isLegacy(src, compressedSize);
(void)dst; (void)dstCapacity; (void)dict; (void)dictSize; /* unused when ZSTD_LEGACY_SUPPORT >= 8 */
switch(version)
{
#if (ZSTD_LEGACY_SUPPORT <= 1)
case 1 :
return ZSTDv01_decompress(dst, dstCapacity, src, compressedSize);
#endif
#if (ZSTD_LEGACY_SUPPORT <= 2)
case 2 :
return ZSTDv02_decompress(dst, dstCapacity, src, compressedSize);
#endif
#if (ZSTD_LEGACY_SUPPORT <= 3)
case 3 :
return ZSTDv03_decompress(dst, dstCapacity, src, compressedSize);
#endif
#if (ZSTD_LEGACY_SUPPORT <= 4)
case 4 :
return ZSTDv04_decompress(dst, dstCapacity, src, compressedSize);
#endif
#if (ZSTD_LEGACY_SUPPORT <= 5)
case 5 :
{ size_t result;
ZSTDv05_DCtx* const zd = ZSTDv05_createDCtx();
if (zd==NULL) return ERROR(memory_allocation);
result = ZSTDv05_decompress_usingDict(zd, dst, dstCapacity, src, compressedSize, dict, dictSize);
ZSTDv05_freeDCtx(zd);
return result;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 6)
case 6 :
{ size_t result;
ZSTDv06_DCtx* const zd = ZSTDv06_createDCtx();
if (zd==NULL) return ERROR(memory_allocation);
result = ZSTDv06_decompress_usingDict(zd, dst, dstCapacity, src, compressedSize, dict, dictSize);
ZSTDv06_freeDCtx(zd);
return result;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 7)
case 7 :
{ size_t result;
ZSTDv07_DCtx* const zd = ZSTDv07_createDCtx();
if (zd==NULL) return ERROR(memory_allocation);
result = ZSTDv07_decompress_usingDict(zd, dst, dstCapacity, src, compressedSize, dict, dictSize);
ZSTDv07_freeDCtx(zd);
return result;
}
#endif
default :
return ERROR(prefix_unknown);
}
}
MEM_STATIC ZSTD_frameSizeInfo ZSTD_findFrameSizeInfoLegacy(const void *src, size_t srcSize)
{
ZSTD_frameSizeInfo frameSizeInfo;
U32 const version = ZSTD_isLegacy(src, srcSize);
switch(version)
{
#if (ZSTD_LEGACY_SUPPORT <= 1)
case 1 :
ZSTDv01_findFrameSizeInfoLegacy(src, srcSize,
&frameSizeInfo.compressedSize,
&frameSizeInfo.decompressedBound);
break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 2)
case 2 :
ZSTDv02_findFrameSizeInfoLegacy(src, srcSize,
&frameSizeInfo.compressedSize,
&frameSizeInfo.decompressedBound);
break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 3)
case 3 :
ZSTDv03_findFrameSizeInfoLegacy(src, srcSize,
&frameSizeInfo.compressedSize,
&frameSizeInfo.decompressedBound);
break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 4)
case 4 :
ZSTDv04_findFrameSizeInfoLegacy(src, srcSize,
&frameSizeInfo.compressedSize,
&frameSizeInfo.decompressedBound);
break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 5)
case 5 :
ZSTDv05_findFrameSizeInfoLegacy(src, srcSize,
&frameSizeInfo.compressedSize,
&frameSizeInfo.decompressedBound);
break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 6)
case 6 :
ZSTDv06_findFrameSizeInfoLegacy(src, srcSize,
&frameSizeInfo.compressedSize,
&frameSizeInfo.decompressedBound);
break;
#endif
#if (ZSTD_LEGACY_SUPPORT <= 7)
case 7 :
ZSTDv07_findFrameSizeInfoLegacy(src, srcSize,
&frameSizeInfo.compressedSize,
&frameSizeInfo.decompressedBound);
break;
#endif
default :
frameSizeInfo.compressedSize = ERROR(prefix_unknown);
frameSizeInfo.decompressedBound = ZSTD_CONTENTSIZE_ERROR;
break;
}
if (!ZSTD_isError(frameSizeInfo.compressedSize) && frameSizeInfo.compressedSize > srcSize) {
frameSizeInfo.compressedSize = ERROR(srcSize_wrong);
frameSizeInfo.decompressedBound = ZSTD_CONTENTSIZE_ERROR;
}
/* In all cases, decompressedBound == nbBlocks * ZSTD_BLOCKSIZE_MAX.
* So we can compute nbBlocks without having to change every function.
*/
if (frameSizeInfo.decompressedBound != ZSTD_CONTENTSIZE_ERROR) {
assert((frameSizeInfo.decompressedBound & (ZSTD_BLOCKSIZE_MAX - 1)) == 0);
frameSizeInfo.nbBlocks = (size_t)(frameSizeInfo.decompressedBound / ZSTD_BLOCKSIZE_MAX);
}
return frameSizeInfo;
}
MEM_STATIC size_t ZSTD_findFrameCompressedSizeLegacy(const void *src, size_t srcSize)
{
ZSTD_frameSizeInfo frameSizeInfo = ZSTD_findFrameSizeInfoLegacy(src, srcSize);
return frameSizeInfo.compressedSize;
}
MEM_STATIC size_t ZSTD_freeLegacyStreamContext(void* legacyContext, U32 version)
{
switch(version)
{
default :
case 1 :
case 2 :
case 3 :
(void)legacyContext;
return ERROR(version_unsupported);
#if (ZSTD_LEGACY_SUPPORT <= 4)
case 4 : return ZBUFFv04_freeDCtx((ZBUFFv04_DCtx*)legacyContext);
#endif
#if (ZSTD_LEGACY_SUPPORT <= 5)
case 5 : return ZBUFFv05_freeDCtx((ZBUFFv05_DCtx*)legacyContext);
#endif
#if (ZSTD_LEGACY_SUPPORT <= 6)
case 6 : return ZBUFFv06_freeDCtx((ZBUFFv06_DCtx*)legacyContext);
#endif
#if (ZSTD_LEGACY_SUPPORT <= 7)
case 7 : return ZBUFFv07_freeDCtx((ZBUFFv07_DCtx*)legacyContext);
#endif
}
}
MEM_STATIC size_t ZSTD_initLegacyStream(void** legacyContext, U32 prevVersion, U32 newVersion,
const void* dict, size_t dictSize)
{
DEBUGLOG(5, "ZSTD_initLegacyStream for v0.%u", newVersion);
if (prevVersion != newVersion) ZSTD_freeLegacyStreamContext(*legacyContext, prevVersion);
switch(newVersion)
{
default :
case 1 :
case 2 :
case 3 :
(void)dict; (void)dictSize;
return 0;
#if (ZSTD_LEGACY_SUPPORT <= 4)
case 4 :
{
ZBUFFv04_DCtx* dctx = (prevVersion != newVersion) ? ZBUFFv04_createDCtx() : (ZBUFFv04_DCtx*)*legacyContext;
if (dctx==NULL) return ERROR(memory_allocation);
ZBUFFv04_decompressInit(dctx);
ZBUFFv04_decompressWithDictionary(dctx, dict, dictSize);
*legacyContext = dctx;
return 0;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 5)
case 5 :
{
ZBUFFv05_DCtx* dctx = (prevVersion != newVersion) ? ZBUFFv05_createDCtx() : (ZBUFFv05_DCtx*)*legacyContext;
if (dctx==NULL) return ERROR(memory_allocation);
ZBUFFv05_decompressInitDictionary(dctx, dict, dictSize);
*legacyContext = dctx;
return 0;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 6)
case 6 :
{
ZBUFFv06_DCtx* dctx = (prevVersion != newVersion) ? ZBUFFv06_createDCtx() : (ZBUFFv06_DCtx*)*legacyContext;
if (dctx==NULL) return ERROR(memory_allocation);
ZBUFFv06_decompressInitDictionary(dctx, dict, dictSize);
*legacyContext = dctx;
return 0;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 7)
case 7 :
{
ZBUFFv07_DCtx* dctx = (prevVersion != newVersion) ? ZBUFFv07_createDCtx() : (ZBUFFv07_DCtx*)*legacyContext;
if (dctx==NULL) return ERROR(memory_allocation);
ZBUFFv07_decompressInitDictionary(dctx, dict, dictSize);
*legacyContext = dctx;
return 0;
}
#endif
}
}
MEM_STATIC size_t ZSTD_decompressLegacyStream(void* legacyContext, U32 version,
ZSTD_outBuffer* output, ZSTD_inBuffer* input)
{
DEBUGLOG(5, "ZSTD_decompressLegacyStream for v0.%u", version);
switch(version)
{
default :
case 1 :
case 2 :
case 3 :
(void)legacyContext; (void)output; (void)input;
return ERROR(version_unsupported);
#if (ZSTD_LEGACY_SUPPORT <= 4)
case 4 :
{
ZBUFFv04_DCtx* dctx = (ZBUFFv04_DCtx*) legacyContext;
const void* src = (const char*)input->src + input->pos;
size_t readSize = input->size - input->pos;
void* dst = (char*)output->dst + output->pos;
size_t decodedSize = output->size - output->pos;
size_t const hintSize = ZBUFFv04_decompressContinue(dctx, dst, &decodedSize, src, &readSize);
output->pos += decodedSize;
input->pos += readSize;
return hintSize;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 5)
case 5 :
{
ZBUFFv05_DCtx* dctx = (ZBUFFv05_DCtx*) legacyContext;
const void* src = (const char*)input->src + input->pos;
size_t readSize = input->size - input->pos;
void* dst = (char*)output->dst + output->pos;
size_t decodedSize = output->size - output->pos;
size_t const hintSize = ZBUFFv05_decompressContinue(dctx, dst, &decodedSize, src, &readSize);
output->pos += decodedSize;
input->pos += readSize;
return hintSize;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 6)
case 6 :
{
ZBUFFv06_DCtx* dctx = (ZBUFFv06_DCtx*) legacyContext;
const void* src = (const char*)input->src + input->pos;
size_t readSize = input->size - input->pos;
void* dst = (char*)output->dst + output->pos;
size_t decodedSize = output->size - output->pos;
size_t const hintSize = ZBUFFv06_decompressContinue(dctx, dst, &decodedSize, src, &readSize);
output->pos += decodedSize;
input->pos += readSize;
return hintSize;
}
#endif
#if (ZSTD_LEGACY_SUPPORT <= 7)
case 7 :
{
ZBUFFv07_DCtx* dctx = (ZBUFFv07_DCtx*) legacyContext;
const void* src = (const char*)input->src + input->pos;
size_t readSize = input->size - input->pos;
void* dst = (char*)output->dst + output->pos;
size_t decodedSize = output->size - output->pos;
size_t const hintSize = ZBUFFv07_decompressContinue(dctx, dst, &decodedSize, src, &readSize);
output->pos += decodedSize;
input->pos += readSize;
return hintSize;
}
#endif
}
}
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_LEGACY_H */

2125
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v01.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

94
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v01.h vendored Normal file
View file

@ -0,0 +1,94 @@
/*
* Copyright (c) Yann Collet, Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_V01_H_28739879432
#define ZSTD_V01_H_28739879432
#if defined (__cplusplus)
extern "C" {
#endif
/* *************************************
* Includes
***************************************/
#include <stddef.h> /* size_t */
/* *************************************
* Simple one-step function
***************************************/
/**
ZSTDv01_decompress() : decompress ZSTD frames compliant with v0.1.x format
compressedSize : is the exact source size
maxOriginalSize : is the size of the 'dst' buffer, which must be already allocated.
It must be equal or larger than originalSize, otherwise decompression will fail.
return : the number of bytes decompressed into destination buffer (originalSize)
or an errorCode if it fails (which can be tested using ZSTDv01_isError())
*/
size_t ZSTDv01_decompress( void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/**
ZSTDv01_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.1.x format
srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
cSize (output parameter) : the number of bytes that would be read to decompress this frame
or an error code if it fails (which can be tested using ZSTDv01_isError())
dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
or ZSTD_CONTENTSIZE_ERROR if an error occurs
note : assumes `cSize` and `dBound` are _not_ NULL.
*/
void ZSTDv01_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
size_t* cSize, unsigned long long* dBound);
/**
ZSTDv01_isError() : tells if the result of ZSTDv01_decompress() is an error
*/
unsigned ZSTDv01_isError(size_t code);
/* *************************************
* Advanced functions
***************************************/
typedef struct ZSTDv01_Dctx_s ZSTDv01_Dctx;
ZSTDv01_Dctx* ZSTDv01_createDCtx(void);
size_t ZSTDv01_freeDCtx(ZSTDv01_Dctx* dctx);
size_t ZSTDv01_decompressDCtx(void* ctx,
void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/* *************************************
* Streaming functions
***************************************/
size_t ZSTDv01_resetDCtx(ZSTDv01_Dctx* dctx);
size_t ZSTDv01_nextSrcSizeToDecompress(ZSTDv01_Dctx* dctx);
size_t ZSTDv01_decompressContinue(ZSTDv01_Dctx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
/**
Use above functions alternatively.
ZSTD_nextSrcSizeToDecompress() tells how much bytes to provide as 'srcSize' to ZSTD_decompressContinue().
ZSTD_decompressContinue() will use previous data blocks to improve compression if they are located prior to current block.
Result is the number of bytes regenerated within 'dst'.
It can be zero, which is not an error; it just means ZSTD_decompressContinue() has decoded some header.
*/
/* *************************************
* Prefix - version detection
***************************************/
#define ZSTDv01_magicNumber 0xFD2FB51E /* Big Endian version */
#define ZSTDv01_magicNumberLE 0x1EB52FFD /* Little Endian version */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_V01_H_28739879432 */

3477
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v02.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

93
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v02.h vendored Normal file
View file

@ -0,0 +1,93 @@
/*
* Copyright (c) Yann Collet, Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_V02_H_4174539423
#define ZSTD_V02_H_4174539423
#if defined (__cplusplus)
extern "C" {
#endif
/* *************************************
* Includes
***************************************/
#include <stddef.h> /* size_t */
/* *************************************
* Simple one-step function
***************************************/
/**
ZSTDv02_decompress() : decompress ZSTD frames compliant with v0.2.x format
compressedSize : is the exact source size
maxOriginalSize : is the size of the 'dst' buffer, which must be already allocated.
It must be equal or larger than originalSize, otherwise decompression will fail.
return : the number of bytes decompressed into destination buffer (originalSize)
or an errorCode if it fails (which can be tested using ZSTDv01_isError())
*/
size_t ZSTDv02_decompress( void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/**
ZSTDv02_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.2.x format
srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
cSize (output parameter) : the number of bytes that would be read to decompress this frame
or an error code if it fails (which can be tested using ZSTDv01_isError())
dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
or ZSTD_CONTENTSIZE_ERROR if an error occurs
note : assumes `cSize` and `dBound` are _not_ NULL.
*/
void ZSTDv02_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
size_t* cSize, unsigned long long* dBound);
/**
ZSTDv02_isError() : tells if the result of ZSTDv02_decompress() is an error
*/
unsigned ZSTDv02_isError(size_t code);
/* *************************************
* Advanced functions
***************************************/
typedef struct ZSTDv02_Dctx_s ZSTDv02_Dctx;
ZSTDv02_Dctx* ZSTDv02_createDCtx(void);
size_t ZSTDv02_freeDCtx(ZSTDv02_Dctx* dctx);
size_t ZSTDv02_decompressDCtx(void* ctx,
void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/* *************************************
* Streaming functions
***************************************/
size_t ZSTDv02_resetDCtx(ZSTDv02_Dctx* dctx);
size_t ZSTDv02_nextSrcSizeToDecompress(ZSTDv02_Dctx* dctx);
size_t ZSTDv02_decompressContinue(ZSTDv02_Dctx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
/**
Use above functions alternatively.
ZSTD_nextSrcSizeToDecompress() tells how much bytes to provide as 'srcSize' to ZSTD_decompressContinue().
ZSTD_decompressContinue() will use previous data blocks to improve compression if they are located prior to current block.
Result is the number of bytes regenerated within 'dst'.
It can be zero, which is not an error; it just means ZSTD_decompressContinue() has decoded some header.
*/
/* *************************************
* Prefix - version detection
***************************************/
#define ZSTDv02_magicNumber 0xFD2FB522 /* v0.2 */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_V02_H_4174539423 */

3117
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v03.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

93
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v03.h vendored Normal file
View file

@ -0,0 +1,93 @@
/*
* Copyright (c) Yann Collet, Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_V03_H_298734209782
#define ZSTD_V03_H_298734209782
#if defined (__cplusplus)
extern "C" {
#endif
/* *************************************
* Includes
***************************************/
#include <stddef.h> /* size_t */
/* *************************************
* Simple one-step function
***************************************/
/**
ZSTDv03_decompress() : decompress ZSTD frames compliant with v0.3.x format
compressedSize : is the exact source size
maxOriginalSize : is the size of the 'dst' buffer, which must be already allocated.
It must be equal or larger than originalSize, otherwise decompression will fail.
return : the number of bytes decompressed into destination buffer (originalSize)
or an errorCode if it fails (which can be tested using ZSTDv01_isError())
*/
size_t ZSTDv03_decompress( void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/**
ZSTDv03_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.3.x format
srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
cSize (output parameter) : the number of bytes that would be read to decompress this frame
or an error code if it fails (which can be tested using ZSTDv01_isError())
dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
or ZSTD_CONTENTSIZE_ERROR if an error occurs
note : assumes `cSize` and `dBound` are _not_ NULL.
*/
void ZSTDv03_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
size_t* cSize, unsigned long long* dBound);
/**
ZSTDv03_isError() : tells if the result of ZSTDv03_decompress() is an error
*/
unsigned ZSTDv03_isError(size_t code);
/* *************************************
* Advanced functions
***************************************/
typedef struct ZSTDv03_Dctx_s ZSTDv03_Dctx;
ZSTDv03_Dctx* ZSTDv03_createDCtx(void);
size_t ZSTDv03_freeDCtx(ZSTDv03_Dctx* dctx);
size_t ZSTDv03_decompressDCtx(void* ctx,
void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/* *************************************
* Streaming functions
***************************************/
size_t ZSTDv03_resetDCtx(ZSTDv03_Dctx* dctx);
size_t ZSTDv03_nextSrcSizeToDecompress(ZSTDv03_Dctx* dctx);
size_t ZSTDv03_decompressContinue(ZSTDv03_Dctx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
/**
Use above functions alternatively.
ZSTD_nextSrcSizeToDecompress() tells how much bytes to provide as 'srcSize' to ZSTD_decompressContinue().
ZSTD_decompressContinue() will use previous data blocks to improve compression if they are located prior to current block.
Result is the number of bytes regenerated within 'dst'.
It can be zero, which is not an error; it just means ZSTD_decompressContinue() has decoded some header.
*/
/* *************************************
* Prefix - version detection
***************************************/
#define ZSTDv03_magicNumber 0xFD2FB523 /* v0.3 */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_V03_H_298734209782 */

3605
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v04.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

142
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v04.h vendored Normal file
View file

@ -0,0 +1,142 @@
/*
* Copyright (c) Yann Collet, Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTD_V04_H_91868324769238
#define ZSTD_V04_H_91868324769238
#if defined (__cplusplus)
extern "C" {
#endif
/* *************************************
* Includes
***************************************/
#include <stddef.h> /* size_t */
/* *************************************
* Simple one-step function
***************************************/
/**
ZSTDv04_decompress() : decompress ZSTD frames compliant with v0.4.x format
compressedSize : is the exact source size
maxOriginalSize : is the size of the 'dst' buffer, which must be already allocated.
It must be equal or larger than originalSize, otherwise decompression will fail.
return : the number of bytes decompressed into destination buffer (originalSize)
or an errorCode if it fails (which can be tested using ZSTDv01_isError())
*/
size_t ZSTDv04_decompress( void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/**
ZSTDv04_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.4.x format
srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
cSize (output parameter) : the number of bytes that would be read to decompress this frame
or an error code if it fails (which can be tested using ZSTDv01_isError())
dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
or ZSTD_CONTENTSIZE_ERROR if an error occurs
note : assumes `cSize` and `dBound` are _not_ NULL.
*/
void ZSTDv04_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
size_t* cSize, unsigned long long* dBound);
/**
ZSTDv04_isError() : tells if the result of ZSTDv04_decompress() is an error
*/
unsigned ZSTDv04_isError(size_t code);
/* *************************************
* Advanced functions
***************************************/
typedef struct ZSTDv04_Dctx_s ZSTDv04_Dctx;
ZSTDv04_Dctx* ZSTDv04_createDCtx(void);
size_t ZSTDv04_freeDCtx(ZSTDv04_Dctx* dctx);
size_t ZSTDv04_decompressDCtx(ZSTDv04_Dctx* dctx,
void* dst, size_t maxOriginalSize,
const void* src, size_t compressedSize);
/* *************************************
* Direct Streaming
***************************************/
size_t ZSTDv04_resetDCtx(ZSTDv04_Dctx* dctx);
size_t ZSTDv04_nextSrcSizeToDecompress(ZSTDv04_Dctx* dctx);
size_t ZSTDv04_decompressContinue(ZSTDv04_Dctx* dctx, void* dst, size_t maxDstSize, const void* src, size_t srcSize);
/**
Use above functions alternatively.
ZSTD_nextSrcSizeToDecompress() tells how much bytes to provide as 'srcSize' to ZSTD_decompressContinue().
ZSTD_decompressContinue() will use previous data blocks to improve compression if they are located prior to current block.
Result is the number of bytes regenerated within 'dst'.
It can be zero, which is not an error; it just means ZSTD_decompressContinue() has decoded some header.
*/
/* *************************************
* Buffered Streaming
***************************************/
typedef struct ZBUFFv04_DCtx_s ZBUFFv04_DCtx;
ZBUFFv04_DCtx* ZBUFFv04_createDCtx(void);
size_t ZBUFFv04_freeDCtx(ZBUFFv04_DCtx* dctx);
size_t ZBUFFv04_decompressInit(ZBUFFv04_DCtx* dctx);
size_t ZBUFFv04_decompressWithDictionary(ZBUFFv04_DCtx* dctx, const void* dict, size_t dictSize);
size_t ZBUFFv04_decompressContinue(ZBUFFv04_DCtx* dctx, void* dst, size_t* maxDstSizePtr, const void* src, size_t* srcSizePtr);
/** ************************************************
* Streaming decompression
*
* A ZBUFF_DCtx object is required to track streaming operation.
* Use ZBUFF_createDCtx() and ZBUFF_freeDCtx() to create/release resources.
* Use ZBUFF_decompressInit() to start a new decompression operation.
* ZBUFF_DCtx objects can be reused multiple times.
*
* Optionally, a reference to a static dictionary can be set, using ZBUFF_decompressWithDictionary()
* It must be the same content as the one set during compression phase.
* Dictionary content must remain accessible during the decompression process.
*
* Use ZBUFF_decompressContinue() repetitively to consume your input.
* *srcSizePtr and *maxDstSizePtr can be any size.
* The function will report how many bytes were read or written by modifying *srcSizePtr and *maxDstSizePtr.
* Note that it may not consume the entire input, in which case it's up to the caller to present remaining input again.
* The content of dst will be overwritten (up to *maxDstSizePtr) at each function call, so save its content if it matters or change dst.
* @return : a hint to preferred nb of bytes to use as input for next function call (it's only a hint, to improve latency)
* or 0 when a frame is completely decoded
* or an error code, which can be tested using ZBUFF_isError().
*
* Hint : recommended buffer sizes (not compulsory) : ZBUFF_recommendedDInSize / ZBUFF_recommendedDOutSize
* output : ZBUFF_recommendedDOutSize==128 KB block size is the internal unit, it ensures it's always possible to write a full block when it's decoded.
* input : ZBUFF_recommendedDInSize==128Kb+3; just follow indications from ZBUFF_decompressContinue() to minimize latency. It should always be <= 128 KB + 3 .
* **************************************************/
unsigned ZBUFFv04_isError(size_t errorCode);
const char* ZBUFFv04_getErrorName(size_t errorCode);
/** The below functions provide recommended buffer sizes for Compression or Decompression operations.
* These sizes are not compulsory, they just tend to offer better latency */
size_t ZBUFFv04_recommendedDInSize(void);
size_t ZBUFFv04_recommendedDOutSize(void);
/* *************************************
* Prefix - version detection
***************************************/
#define ZSTDv04_magicNumber 0xFD2FB524 /* v0.4 */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTD_V04_H_91868324769238 */

4004
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v05.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

162
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v05.h vendored Normal file
View file

@ -0,0 +1,162 @@
/*
* Copyright (c) Yann Collet, Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTDv05_H
#define ZSTDv05_H
#if defined (__cplusplus)
extern "C" {
#endif
/*-*************************************
* Dependencies
***************************************/
#include <stddef.h> /* size_t */
#include "../common/mem.h" /* U64, U32 */
/* *************************************
* Simple functions
***************************************/
/*! ZSTDv05_decompress() :
`compressedSize` : is the _exact_ size of the compressed blob, otherwise decompression will fail.
`dstCapacity` must be large enough, equal or larger than originalSize.
@return : the number of bytes decompressed into `dst` (<= `dstCapacity`),
or an errorCode if it fails (which can be tested using ZSTDv05_isError()) */
size_t ZSTDv05_decompress( void* dst, size_t dstCapacity,
const void* src, size_t compressedSize);
/**
ZSTDv05_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.5.x format
srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
cSize (output parameter) : the number of bytes that would be read to decompress this frame
or an error code if it fails (which can be tested using ZSTDv01_isError())
dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
or ZSTD_CONTENTSIZE_ERROR if an error occurs
note : assumes `cSize` and `dBound` are _not_ NULL.
*/
void ZSTDv05_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
size_t* cSize, unsigned long long* dBound);
/* *************************************
* Helper functions
***************************************/
/* Error Management */
unsigned ZSTDv05_isError(size_t code); /*!< tells if a `size_t` function result is an error code */
const char* ZSTDv05_getErrorName(size_t code); /*!< provides readable string for an error code */
/* *************************************
* Explicit memory management
***************************************/
/** Decompression context */
typedef struct ZSTDv05_DCtx_s ZSTDv05_DCtx;
ZSTDv05_DCtx* ZSTDv05_createDCtx(void);
size_t ZSTDv05_freeDCtx(ZSTDv05_DCtx* dctx); /*!< @return : errorCode */
/** ZSTDv05_decompressDCtx() :
* Same as ZSTDv05_decompress(), but requires an already allocated ZSTDv05_DCtx (see ZSTDv05_createDCtx()) */
size_t ZSTDv05_decompressDCtx(ZSTDv05_DCtx* ctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/*-***********************
* Simple Dictionary API
*************************/
/*! ZSTDv05_decompress_usingDict() :
* Decompression using a pre-defined Dictionary content (see dictBuilder).
* Dictionary must be identical to the one used during compression, otherwise regenerated data will be corrupted.
* Note : dict can be NULL, in which case, it's equivalent to ZSTDv05_decompressDCtx() */
size_t ZSTDv05_decompress_usingDict(ZSTDv05_DCtx* dctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const void* dict,size_t dictSize);
/*-************************
* Advanced Streaming API
***************************/
typedef enum { ZSTDv05_fast, ZSTDv05_greedy, ZSTDv05_lazy, ZSTDv05_lazy2, ZSTDv05_btlazy2, ZSTDv05_opt, ZSTDv05_btopt } ZSTDv05_strategy;
typedef struct {
U64 srcSize;
U32 windowLog; /* the only useful information to retrieve */
U32 contentLog; U32 hashLog; U32 searchLog; U32 searchLength; U32 targetLength; ZSTDv05_strategy strategy;
} ZSTDv05_parameters;
size_t ZSTDv05_getFrameParams(ZSTDv05_parameters* params, const void* src, size_t srcSize);
size_t ZSTDv05_decompressBegin_usingDict(ZSTDv05_DCtx* dctx, const void* dict, size_t dictSize);
void ZSTDv05_copyDCtx(ZSTDv05_DCtx* dstDCtx, const ZSTDv05_DCtx* srcDCtx);
size_t ZSTDv05_nextSrcSizeToDecompress(ZSTDv05_DCtx* dctx);
size_t ZSTDv05_decompressContinue(ZSTDv05_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/*-***********************
* ZBUFF API
*************************/
typedef struct ZBUFFv05_DCtx_s ZBUFFv05_DCtx;
ZBUFFv05_DCtx* ZBUFFv05_createDCtx(void);
size_t ZBUFFv05_freeDCtx(ZBUFFv05_DCtx* dctx);
size_t ZBUFFv05_decompressInit(ZBUFFv05_DCtx* dctx);
size_t ZBUFFv05_decompressInitDictionary(ZBUFFv05_DCtx* dctx, const void* dict, size_t dictSize);
size_t ZBUFFv05_decompressContinue(ZBUFFv05_DCtx* dctx,
void* dst, size_t* dstCapacityPtr,
const void* src, size_t* srcSizePtr);
/*-***************************************************************************
* Streaming decompression
*
* A ZBUFFv05_DCtx object is required to track streaming operations.
* Use ZBUFFv05_createDCtx() and ZBUFFv05_freeDCtx() to create/release resources.
* Use ZBUFFv05_decompressInit() to start a new decompression operation,
* or ZBUFFv05_decompressInitDictionary() if decompression requires a dictionary.
* Note that ZBUFFv05_DCtx objects can be reused multiple times.
*
* Use ZBUFFv05_decompressContinue() repetitively to consume your input.
* *srcSizePtr and *dstCapacityPtr can be any size.
* The function will report how many bytes were read or written by modifying *srcSizePtr and *dstCapacityPtr.
* Note that it may not consume the entire input, in which case it's up to the caller to present remaining input again.
* The content of @dst will be overwritten (up to *dstCapacityPtr) at each function call, so save its content if it matters or change @dst.
* @return : a hint to preferred nb of bytes to use as input for next function call (it's only a hint, to help latency)
* or 0 when a frame is completely decoded
* or an error code, which can be tested using ZBUFFv05_isError().
*
* Hint : recommended buffer sizes (not compulsory) : ZBUFFv05_recommendedDInSize() / ZBUFFv05_recommendedDOutSize()
* output : ZBUFFv05_recommendedDOutSize==128 KB block size is the internal unit, it ensures it's always possible to write a full block when decoded.
* input : ZBUFFv05_recommendedDInSize==128Kb+3; just follow indications from ZBUFFv05_decompressContinue() to minimize latency. It should always be <= 128 KB + 3 .
* *******************************************************************************/
/* *************************************
* Tool functions
***************************************/
unsigned ZBUFFv05_isError(size_t errorCode);
const char* ZBUFFv05_getErrorName(size_t errorCode);
/** Functions below provide recommended buffer sizes for Compression or Decompression operations.
* These sizes are just hints, and tend to offer better latency */
size_t ZBUFFv05_recommendedDInSize(void);
size_t ZBUFFv05_recommendedDOutSize(void);
/*-*************************************
* Constants
***************************************/
#define ZSTDv05_MAGICNUMBER 0xFD2FB525 /* v0.5 */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTDv0505_H */

4113
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v06.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

172
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v06.h vendored Normal file
View file

@ -0,0 +1,172 @@
/*
* Copyright (c) Yann Collet, Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTDv06_H
#define ZSTDv06_H
#if defined (__cplusplus)
extern "C" {
#endif
/*====== Dependency ======*/
#include <stddef.h> /* size_t */
/*====== Export for Windows ======*/
/*!
* ZSTDv06_DLL_EXPORT :
* Enable exporting of functions when building a Windows DLL
*/
#if defined(_WIN32) && defined(ZSTDv06_DLL_EXPORT) && (ZSTDv06_DLL_EXPORT==1)
# define ZSTDLIBv06_API __declspec(dllexport)
#else
# define ZSTDLIBv06_API
#endif
/* *************************************
* Simple functions
***************************************/
/*! ZSTDv06_decompress() :
`compressedSize` : is the _exact_ size of the compressed blob, otherwise decompression will fail.
`dstCapacity` must be large enough, equal or larger than originalSize.
@return : the number of bytes decompressed into `dst` (<= `dstCapacity`),
or an errorCode if it fails (which can be tested using ZSTDv06_isError()) */
ZSTDLIBv06_API size_t ZSTDv06_decompress( void* dst, size_t dstCapacity,
const void* src, size_t compressedSize);
/**
ZSTDv06_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.6.x format
srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
cSize (output parameter) : the number of bytes that would be read to decompress this frame
or an error code if it fails (which can be tested using ZSTDv01_isError())
dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
or ZSTD_CONTENTSIZE_ERROR if an error occurs
note : assumes `cSize` and `dBound` are _not_ NULL.
*/
void ZSTDv06_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
size_t* cSize, unsigned long long* dBound);
/* *************************************
* Helper functions
***************************************/
ZSTDLIBv06_API size_t ZSTDv06_compressBound(size_t srcSize); /*!< maximum compressed size (worst case scenario) */
/* Error Management */
ZSTDLIBv06_API unsigned ZSTDv06_isError(size_t code); /*!< tells if a `size_t` function result is an error code */
ZSTDLIBv06_API const char* ZSTDv06_getErrorName(size_t code); /*!< provides readable string for an error code */
/* *************************************
* Explicit memory management
***************************************/
/** Decompression context */
typedef struct ZSTDv06_DCtx_s ZSTDv06_DCtx;
ZSTDLIBv06_API ZSTDv06_DCtx* ZSTDv06_createDCtx(void);
ZSTDLIBv06_API size_t ZSTDv06_freeDCtx(ZSTDv06_DCtx* dctx); /*!< @return : errorCode */
/** ZSTDv06_decompressDCtx() :
* Same as ZSTDv06_decompress(), but requires an already allocated ZSTDv06_DCtx (see ZSTDv06_createDCtx()) */
ZSTDLIBv06_API size_t ZSTDv06_decompressDCtx(ZSTDv06_DCtx* ctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/*-***********************
* Dictionary API
*************************/
/*! ZSTDv06_decompress_usingDict() :
* Decompression using a pre-defined Dictionary content (see dictBuilder).
* Dictionary must be identical to the one used during compression, otherwise regenerated data will be corrupted.
* Note : dict can be NULL, in which case, it's equivalent to ZSTDv06_decompressDCtx() */
ZSTDLIBv06_API size_t ZSTDv06_decompress_usingDict(ZSTDv06_DCtx* dctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const void* dict,size_t dictSize);
/*-************************
* Advanced Streaming API
***************************/
struct ZSTDv06_frameParams_s { unsigned long long frameContentSize; unsigned windowLog; };
typedef struct ZSTDv06_frameParams_s ZSTDv06_frameParams;
ZSTDLIBv06_API size_t ZSTDv06_getFrameParams(ZSTDv06_frameParams* fparamsPtr, const void* src, size_t srcSize); /**< doesn't consume input */
ZSTDLIBv06_API size_t ZSTDv06_decompressBegin_usingDict(ZSTDv06_DCtx* dctx, const void* dict, size_t dictSize);
ZSTDLIBv06_API void ZSTDv06_copyDCtx(ZSTDv06_DCtx* dctx, const ZSTDv06_DCtx* preparedDCtx);
ZSTDLIBv06_API size_t ZSTDv06_nextSrcSizeToDecompress(ZSTDv06_DCtx* dctx);
ZSTDLIBv06_API size_t ZSTDv06_decompressContinue(ZSTDv06_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/* *************************************
* ZBUFF API
***************************************/
typedef struct ZBUFFv06_DCtx_s ZBUFFv06_DCtx;
ZSTDLIBv06_API ZBUFFv06_DCtx* ZBUFFv06_createDCtx(void);
ZSTDLIBv06_API size_t ZBUFFv06_freeDCtx(ZBUFFv06_DCtx* dctx);
ZSTDLIBv06_API size_t ZBUFFv06_decompressInit(ZBUFFv06_DCtx* dctx);
ZSTDLIBv06_API size_t ZBUFFv06_decompressInitDictionary(ZBUFFv06_DCtx* dctx, const void* dict, size_t dictSize);
ZSTDLIBv06_API size_t ZBUFFv06_decompressContinue(ZBUFFv06_DCtx* dctx,
void* dst, size_t* dstCapacityPtr,
const void* src, size_t* srcSizePtr);
/*-***************************************************************************
* Streaming decompression howto
*
* A ZBUFFv06_DCtx object is required to track streaming operations.
* Use ZBUFFv06_createDCtx() and ZBUFFv06_freeDCtx() to create/release resources.
* Use ZBUFFv06_decompressInit() to start a new decompression operation,
* or ZBUFFv06_decompressInitDictionary() if decompression requires a dictionary.
* Note that ZBUFFv06_DCtx objects can be re-init multiple times.
*
* Use ZBUFFv06_decompressContinue() repetitively to consume your input.
* *srcSizePtr and *dstCapacityPtr can be any size.
* The function will report how many bytes were read or written by modifying *srcSizePtr and *dstCapacityPtr.
* Note that it may not consume the entire input, in which case it's up to the caller to present remaining input again.
* The content of `dst` will be overwritten (up to *dstCapacityPtr) at each function call, so save its content if it matters, or change `dst`.
* @return : a hint to preferred nb of bytes to use as input for next function call (it's only a hint, to help latency),
* or 0 when a frame is completely decoded,
* or an error code, which can be tested using ZBUFFv06_isError().
*
* Hint : recommended buffer sizes (not compulsory) : ZBUFFv06_recommendedDInSize() and ZBUFFv06_recommendedDOutSize()
* output : ZBUFFv06_recommendedDOutSize== 128 KB block size is the internal unit, it ensures it's always possible to write a full block when decoded.
* input : ZBUFFv06_recommendedDInSize == 128KB + 3;
* just follow indications from ZBUFFv06_decompressContinue() to minimize latency. It should always be <= 128 KB + 3 .
* *******************************************************************************/
/* *************************************
* Tool functions
***************************************/
ZSTDLIBv06_API unsigned ZBUFFv06_isError(size_t errorCode);
ZSTDLIBv06_API const char* ZBUFFv06_getErrorName(size_t errorCode);
/** Functions below provide recommended buffer sizes for Compression or Decompression operations.
* These sizes are just hints, they tend to offer better latency */
ZSTDLIBv06_API size_t ZBUFFv06_recommendedDInSize(void);
ZSTDLIBv06_API size_t ZBUFFv06_recommendedDOutSize(void);
/*-*************************************
* Constants
***************************************/
#define ZSTDv06_MAGICNUMBER 0xFD2FB526 /* v0.6 */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTDv06_BUFFERED_H */

4498
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v07.c vendored Normal file
View file

File diff suppressed because it is too large Load diff

187
raytracer/nvpro_core/third_party/zstd/lib/legacy/zstd_v07.h vendored Normal file
View file

@ -0,0 +1,187 @@
/*
* Copyright (c) Yann Collet, Meta Platforms, Inc. and affiliates.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
#ifndef ZSTDv07_H_235446
#define ZSTDv07_H_235446
#if defined (__cplusplus)
extern "C" {
#endif
/*====== Dependency ======*/
#include <stddef.h> /* size_t */
/*====== Export for Windows ======*/
/*!
* ZSTDv07_DLL_EXPORT :
* Enable exporting of functions when building a Windows DLL
*/
#if defined(_WIN32) && defined(ZSTDv07_DLL_EXPORT) && (ZSTDv07_DLL_EXPORT==1)
# define ZSTDLIBv07_API __declspec(dllexport)
#else
# define ZSTDLIBv07_API
#endif
/* *************************************
* Simple API
***************************************/
/*! ZSTDv07_getDecompressedSize() :
* @return : decompressed size if known, 0 otherwise.
note 1 : if `0`, follow up with ZSTDv07_getFrameParams() to know precise failure cause.
note 2 : decompressed size could be wrong or intentionally modified !
always ensure results fit within application's authorized limits */
unsigned long long ZSTDv07_getDecompressedSize(const void* src, size_t srcSize);
/*! ZSTDv07_decompress() :
`compressedSize` : must be _exact_ size of compressed input, otherwise decompression will fail.
`dstCapacity` must be equal or larger than originalSize.
@return : the number of bytes decompressed into `dst` (<= `dstCapacity`),
or an errorCode if it fails (which can be tested using ZSTDv07_isError()) */
ZSTDLIBv07_API size_t ZSTDv07_decompress( void* dst, size_t dstCapacity,
const void* src, size_t compressedSize);
/**
ZSTDv07_findFrameSizeInfoLegacy() : get the source length and decompressed bound of a ZSTD frame compliant with v0.7.x format
srcSize : The size of the 'src' buffer, at least as large as the frame pointed to by 'src'
cSize (output parameter) : the number of bytes that would be read to decompress this frame
or an error code if it fails (which can be tested using ZSTDv01_isError())
dBound (output parameter) : an upper-bound for the decompressed size of the data in the frame
or ZSTD_CONTENTSIZE_ERROR if an error occurs
note : assumes `cSize` and `dBound` are _not_ NULL.
*/
void ZSTDv07_findFrameSizeInfoLegacy(const void *src, size_t srcSize,
size_t* cSize, unsigned long long* dBound);
/*====== Helper functions ======*/
ZSTDLIBv07_API unsigned ZSTDv07_isError(size_t code); /*!< tells if a `size_t` function result is an error code */
ZSTDLIBv07_API const char* ZSTDv07_getErrorName(size_t code); /*!< provides readable string from an error code */
/*-*************************************
* Explicit memory management
***************************************/
/** Decompression context */
typedef struct ZSTDv07_DCtx_s ZSTDv07_DCtx;
ZSTDLIBv07_API ZSTDv07_DCtx* ZSTDv07_createDCtx(void);
ZSTDLIBv07_API size_t ZSTDv07_freeDCtx(ZSTDv07_DCtx* dctx); /*!< @return : errorCode */
/** ZSTDv07_decompressDCtx() :
* Same as ZSTDv07_decompress(), requires an allocated ZSTDv07_DCtx (see ZSTDv07_createDCtx()) */
ZSTDLIBv07_API size_t ZSTDv07_decompressDCtx(ZSTDv07_DCtx* ctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
/*-************************
* Simple dictionary API
***************************/
/*! ZSTDv07_decompress_usingDict() :
* Decompression using a pre-defined Dictionary content (see dictBuilder).
* Dictionary must be identical to the one used during compression.
* Note : This function load the dictionary, resulting in a significant startup time */
ZSTDLIBv07_API size_t ZSTDv07_decompress_usingDict(ZSTDv07_DCtx* dctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const void* dict,size_t dictSize);
/*-**************************
* Advanced Dictionary API
****************************/
/*! ZSTDv07_createDDict() :
* Create a digested dictionary, ready to start decompression operation without startup delay.
* `dict` can be released after creation */
typedef struct ZSTDv07_DDict_s ZSTDv07_DDict;
ZSTDLIBv07_API ZSTDv07_DDict* ZSTDv07_createDDict(const void* dict, size_t dictSize);
ZSTDLIBv07_API size_t ZSTDv07_freeDDict(ZSTDv07_DDict* ddict);
/*! ZSTDv07_decompress_usingDDict() :
* Decompression using a pre-digested Dictionary
* Faster startup than ZSTDv07_decompress_usingDict(), recommended when same dictionary is used multiple times. */
ZSTDLIBv07_API size_t ZSTDv07_decompress_usingDDict(ZSTDv07_DCtx* dctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const ZSTDv07_DDict* ddict);
typedef struct {
unsigned long long frameContentSize;
unsigned windowSize;
unsigned dictID;
unsigned checksumFlag;
} ZSTDv07_frameParams;
ZSTDLIBv07_API size_t ZSTDv07_getFrameParams(ZSTDv07_frameParams* fparamsPtr, const void* src, size_t srcSize); /**< doesn't consume input */
/* *************************************
* Streaming functions
***************************************/
typedef struct ZBUFFv07_DCtx_s ZBUFFv07_DCtx;
ZSTDLIBv07_API ZBUFFv07_DCtx* ZBUFFv07_createDCtx(void);
ZSTDLIBv07_API size_t ZBUFFv07_freeDCtx(ZBUFFv07_DCtx* dctx);
ZSTDLIBv07_API size_t ZBUFFv07_decompressInit(ZBUFFv07_DCtx* dctx);
ZSTDLIBv07_API size_t ZBUFFv07_decompressInitDictionary(ZBUFFv07_DCtx* dctx, const void* dict, size_t dictSize);
ZSTDLIBv07_API size_t ZBUFFv07_decompressContinue(ZBUFFv07_DCtx* dctx,
void* dst, size_t* dstCapacityPtr,
const void* src, size_t* srcSizePtr);
/*-***************************************************************************
* Streaming decompression howto
*
* A ZBUFFv07_DCtx object is required to track streaming operations.
* Use ZBUFFv07_createDCtx() and ZBUFFv07_freeDCtx() to create/release resources.
* Use ZBUFFv07_decompressInit() to start a new decompression operation,
* or ZBUFFv07_decompressInitDictionary() if decompression requires a dictionary.
* Note that ZBUFFv07_DCtx objects can be re-init multiple times.
*
* Use ZBUFFv07_decompressContinue() repetitively to consume your input.
* *srcSizePtr and *dstCapacityPtr can be any size.
* The function will report how many bytes were read or written by modifying *srcSizePtr and *dstCapacityPtr.
* Note that it may not consume the entire input, in which case it's up to the caller to present remaining input again.
* The content of `dst` will be overwritten (up to *dstCapacityPtr) at each function call, so save its content if it matters, or change `dst`.
* @return : a hint to preferred nb of bytes to use as input for next function call (it's only a hint, to help latency),
* or 0 when a frame is completely decoded,
* or an error code, which can be tested using ZBUFFv07_isError().
*
* Hint : recommended buffer sizes (not compulsory) : ZBUFFv07_recommendedDInSize() and ZBUFFv07_recommendedDOutSize()
* output : ZBUFFv07_recommendedDOutSize== 128 KB block size is the internal unit, it ensures it's always possible to write a full block when decoded.
* input : ZBUFFv07_recommendedDInSize == 128KB + 3;
* just follow indications from ZBUFFv07_decompressContinue() to minimize latency. It should always be <= 128 KB + 3 .
* *******************************************************************************/
/* *************************************
* Tool functions
***************************************/
ZSTDLIBv07_API unsigned ZBUFFv07_isError(size_t errorCode);
ZSTDLIBv07_API const char* ZBUFFv07_getErrorName(size_t errorCode);
/** Functions below provide recommended buffer sizes for Compression or Decompression operations.
* These sizes are just hints, they tend to offer better latency */
ZSTDLIBv07_API size_t ZBUFFv07_recommendedDInSize(void);
ZSTDLIBv07_API size_t ZBUFFv07_recommendedDOutSize(void);
/*-*************************************
* Constants
***************************************/
#define ZSTDv07_MAGICNUMBER 0xFD2FB527 /* v0.7 */
#if defined (__cplusplus)
}
#endif
#endif /* ZSTDv07_H_235446 */

16
raytracer/nvpro_core/third_party/zstd/lib/libzstd.pc.in vendored Normal file
View file

@ -0,0 +1,16 @@
# ZSTD - standard compression algorithm
# Copyright (c) Meta Platforms, Inc. and affiliates.
# BSD 2-Clause License (https://opensource.org/licenses/bsd-license.php)
prefix=@PREFIX@
exec_prefix=@EXEC_PREFIX@
includedir=@INCLUDEDIR@
libdir=@LIBDIR@
Name: zstd
Description: fast lossless compression algorithm library
URL: https://facebook.github.io/zstd/
Version: @VERSION@
Libs: -L${libdir} -lzstd
Libs.private: @LIBS_PRIVATE@
Cflags: -I${includedir}

Some files were not shown because too many files have changed in this diff Show more