9c862c6a53
Fix manpage symlinks [skip ci]
2018-12-01 23:18:59 +07:00
d79df2a370
Apply new InstallSymlink script
2018-12-01 23:18:59 +07:00
ef2e761937
Helper script to install symlink in meson
2018-12-01 23:18:59 +07:00
3175188407
No need these helpers
2018-12-01 23:18:59 +07:00
337f914dc8
Fix lib soversion and no install cover.h header
2018-12-01 23:18:59 +07:00
c9f0144302
Fix meson tests build
2018-12-01 23:18:59 +07:00
5a36a57cf5
Bump to 1.3.8 and fix run_command function
...
The run_command is run from an unspecified directory. Therefore we cannot assume
which directory it is running our command.
2018-12-01 23:18:59 +07:00
8a160680d1
Update legacy support to 5
2018-12-01 23:18:59 +07:00
f727808731
Minor fix for meson build
...
Use files function instead of constructing path with meson.current_source_dir()
2018-12-01 23:18:59 +07:00
9a721e5216
Update meson build system
...
NOTE: This commit only tested on Linux (Ubuntu 18.04). Windows
build may not work as expected.
* Use meson >= 0.47.0 cause we use install_man function
* Add three helper Python script:
* CopyFile.py: To copy file
* CreateSymlink.py: To make symlink (both Windows and Unix)
* GetZstdLibraryVersion.py: Parse lib/zstd.h to get zstd version
These help emulating equivalent functions in CMake and Makefile.
* Use subdir from meson to split meson.build
* Add contrib build
* Fix other build
* Add new build options
* build_programs: Enable programs build
* build_contrib: Enable contrib build
* build_tests: Enable tests build
* use_static_runtime: Link to static run-time libraries on MSVC
* zlib_support: Enable zlib support
* lzma_support: Enable lzma support
2018-11-28 01:08:34 +07:00
9bd8f6a00c
Rename and update build instruction in README file to README.md
2018-11-28 01:08:34 +07:00
2abd5139a5
Add meson build guide
2018-11-28 01:08:34 +07:00
5adbad4059
Merge branch 'dev' into advancedAPI
2018-11-14 13:00:37 -08:00
b83d1e7714
removed some `static const` variables
...
and replaced by traditional macro constants.
Unfortunately, C doesn't consider `static const` to mean "constant"
2018-11-13 16:56:32 -08:00
b830ccca5c
changed benchfn api
...
to use structure for function parameters
as it expresses much clearer than a long list of parameters,
since each parameter can now be named.
2018-11-13 13:12:50 -08:00
d38063f8ae
separated bench module into benchfn and benchzstd
...
it shall be possible to use benchfn
without any dependency on zstd.
2018-11-13 11:01:59 -08:00
483759a3de
Improves decompression speed when using cold dictionary
...
by triggering the prefetching decoder path
(which used to be dedicated to long-range offsets only).
Figures on my laptop :
no content prefetch : ~300 MB/s (for reference)
full content prefetch : ~325 MB/s (before this patch)
new prefetch path : ~375 MB/s (after this patch)
The benchmark speed is already significant,
but another side-effect is that this version
prefetch less data into memory,
since it only prefetches what's needed, instead of the full dictionary.
This is supposed to help highly active environments
such as active databases,
that can't be properly measured in benchmark environment (too clean).
Also :
fixed the largeNbDict test program
which was working improperly when setting nbBlocks > nbFiles.
2018-11-08 17:00:23 -08:00
705e0b18ab
Making changes to make it compile on my laptop
2018-10-11 15:51:57 -07:00
123fac6b6d
fix pzstd compatibility with mingw
...
some details changed with introduction of gcc7
2018-09-21 17:36:00 -07:00
00ce26725b
Merge pull request #1324 from ko-zu/fixclangcode
...
Fix largeNbDicts bench for clangbuild
2018-09-17 14:10:17 -07:00
8f27e8cf3d
Merge pull request #1322 from azat-archive/seekable-fixes-pull
...
Fixes read write past end of input buffer.
2018-09-17 11:04:51 -07:00
b053bec2f4
Fix largeNbDicts bench for clangbuild
...
Remove unsigned to size_t promotion to fix implicit down conversion errors in clangbuild target.
2018-09-17 13:09:08 +09:00
d707692e05
seekable_decompression: support offset greater then UNIT_MAX
2018-09-16 18:05:32 +03:00
b52867a97f
zstdseek_decompress: fix decompression with data left in input buffer
2018-09-16 18:05:32 +03:00
c49ccbc8e7
largeNbDicts : can select a nb of blocks
...
will automatically truncate or repeat input as needed,
to create the requested nb of blocks.
default: nb of files, eventually increased appropriately if blockSize is set
2018-09-12 11:31:28 -07:00
50b216146f
Merge pull request #1304 from facebook/largeNbDicts
...
contrib/largeNbDicts
2018-09-06 09:50:56 -07:00
c57a856d64
fixed minor static analyzer warning
2018-09-05 14:33:51 -07:00
1d487d587f
updated documentation
2018-09-04 14:57:45 -07:00
11b8b8c100
silenced false-positive scan-build warning
2018-08-31 10:01:06 -07:00
0ff67511e6
fixed link order for old compilers
2018-08-30 16:43:28 -07:00
f76253bb70
minor : createDictionaryBuffer() can create dictionaries of different sizes
2018-08-30 16:24:44 -07:00
39c55a118f
fixed minor compatibility issues with older compilers
2018-08-30 16:00:57 -07:00
39ef91a599
-std=c99 for largeNbDicts
2018-08-30 14:59:23 -07:00
4086b2871b
largeNbDicts compatible with multiple source files
...
splitting is disabled by default, but can be re-enabled using usual command -B#
update commands to look like zstd ones
2018-08-30 14:38:49 -07:00
a5a77965d3
make all includes contrib/largeNbDicts
2018-08-29 16:17:22 -07:00
d89fa814c1
added a README
...
for documentation
2018-08-28 18:19:19 -07:00
6444c50035
increases randomness of ddict ptrs
2018-08-28 18:13:46 -07:00
6c398df241
level, block size and nb dicts can be set on command line
2018-08-28 18:05:31 -07:00
0c66a44d1b
first working test program
...
measures :
- compression ratio with / without dictionary
- create one dictionary per block
- memory budget for dictionaries
- decompression speed, using one different dictionary per block
current limitations :
- only one file
- 4K blocks only
- automatic dictionary built with 4K size
dictionary can be selected on command line, with -D
2018-08-28 15:47:07 -07:00
274b60e6e6
largeNbDicts can compress and compare dict vs noDict
2018-08-27 17:08:44 -07:00
6782725155
first sketch for largeNbDicts test program
2018-08-26 19:29:12 -07:00
9d6ed9def3
Merge fastCover into DictBuilder ( #1274 )
...
* Minor fix
* Run non-optimize FASTCOVER 5 times in benchmark
* Merge fastCover into dictBuilder
* Fix mixed declaration issue
* Add fastcover to symbol.c
* Add fastCover.c and cover.h to build
* Change fastCover.c to fastcover.c
* Update benchmark to run FASTCOVER in dictBuilder
* Undo spliting fastcover_param into cover_param and f
* Remove convert param functions
* Assign f to parameter
* Add zdict.h to Makefile in lib
* Add cover.h to BUCK
* Cast 1 to U64 before shifting
* Remove trimming of zero freq head and tail in selectSegment and rebenchmark
* Remove f as a separate parameter of tryParam
* Read 8 bytes when d is 6
* Add trimming off zero frequency head and tail
* Use best functions from COVER and remove trimming part(which leads to worse compression ratio after previous bugs were fixed)
* Add finalize= argument to FASTCOVER to specify percentage of training samples passed to ZDICT_finalizeDictionary
* Change nbDmer to always read 8 bytes even when d=6
* Add skip=# argument to allow skipping dmers in computeFrequency in FASTCOVER
* Update comments and benchmarking result
* Change default method of ZDICT_trainFromBuffer to ZDICT_optimizeTrainFromBuffer_fastCover
* Add dictType enum and fix bug about passing zParam when converting to coverParam
* Combine finalize and skip into a single parameter
* Update acceleration parameters and benchmark on 3 sample sets
* Change default splitPoint of FASTCOVER to 0.75 and benchmark first 3 sample sets
* Initialize variables outside of for loop in benchmark.c
* Update benchmark result for hg-manifest
* Remove cover.h from install-includes
* Add explanation of f
* Set default compression level for trainFromBuffer to 3
* Add assertion of fastCoverParams in DiB_trainFromFiles
* Add checkTotalCompressedSize function + some minor fixes
* Add test for multithreading fastCovr
* Initialize segmentFreqs in every FASTCOVER_selectSegment and move mutex_unnlock to end of COVER_best_finish
* Free segmentFreqs
* Initialize segmentFreqs before calling FASTCOVER_buildDictionary instead of in FASTCOVER_selectSegment
* Add FASTCOVER_MEMMULT
* Minor fix
* Update benchmarking result
2018-08-23 12:06:20 -07:00
36d6165a2d
Makefile: added variable SCANBUILD
...
so that a different version of scan-build can be selected
2018-08-16 16:44:13 -07:00
42a02ab745
fixed minor warnings issued by scan-build
2018-08-15 14:36:02 -07:00
0acb0abd1e
Add non-optimize FASTCOVER ( #1260 )
...
* Add non-optimize FASTCOVER
* Minor fix
* Pass param as value instead of pointer
2018-08-01 11:06:16 -07:00
4e29bc2469
Use CDict instead of CCtx in analyzeEntropy
2018-07-31 10:36:45 -07:00
31229e527b
Increment frequency for every dmer occurence within same sample instead of at most once per sample
2018-07-30 12:54:22 -07:00
51b109c1b5
Delete old benchmarking result
2018-07-27 17:31:33 -07:00
53ef22a4bc
Undo deleting clean in make
2018-07-27 16:56:50 -07:00
96d84ee235
Revert test.sh
2018-07-27 16:54:05 -07:00
61262f6c0d
Save segmentFreqs in ctx instead of malloc and memset in SelectSegment
2018-07-27 16:51:38 -07:00
49b398e93f
Use same param after optimizing cover and fastCover and record k and d for benchmarking
2018-07-27 13:39:19 -07:00
759c543312
Rerun cover and fastCover with optimized values
2018-07-26 19:03:01 -07:00
3d7941ce41
Benchmark different f values
2018-07-26 16:24:13 -07:00
3b163e0b5b
Add array to keep track of frequency within active segment, fix malloc bug, update benchmarking result
2018-07-26 13:53:13 -07:00
2333ecb173
Allow d=6
2018-07-25 18:10:09 -07:00
1e85f314d8
Benchmark fast cover optimize vs k=200
2018-07-25 17:53:38 -07:00
d1fc507ef9
Initial benchmarking result for fastCover
2018-07-25 17:05:54 -07:00
f5407e398a
Make hash value const
2018-07-25 16:54:08 -07:00
7f3f70f766
Add Fast Cover Dictionary Builder
2018-07-25 16:34:07 -07:00
77068a8447
Merge pull request #1246 from jennifermliu/benchmark
...
Benchmark dictionary builders
2018-07-20 18:09:31 -07:00
b6c5d4982c
Minor fix
2018-07-20 17:41:22 -07:00
71e767ac09
Refactoring and benchmark without dictionary
2018-07-20 17:03:47 -07:00
470c8d42f4
Benchmark dictionary builders
2018-07-20 11:32:39 -07:00
4d1ad5cdb2
Merge pull request #1238 from jennifermliu/random
...
Add random dictionary builder
2018-07-19 13:52:15 -07:00
0c5eaef248
Update Makefile
2018-07-19 13:44:27 -07:00
5bb46a898e
Rename cleanup
2018-07-18 12:15:49 -07:00
52e7cf0e40
Add cleanup to trainfromFiles and move RANDOM_segment_t declaration
2018-07-18 10:40:13 -07:00
ce09fb723d
Update freeSampleInfo
2018-07-17 16:13:40 -07:00
896ff0644a
Fix deallocation problem and add documentation
2018-07-17 16:01:44 -07:00
e6fe405838
Make test PHONY target
2018-07-17 12:42:53 -07:00
49acfaeaec
Move file loading functions to new file for access by benchmarking tool
2018-07-17 12:35:09 -07:00
4d32339b75
Remove CLevel cli option which was accidentally added back in the last commit
2018-07-16 18:59:18 -07:00
1f7fa5cdd6
Fix spacing and Edit Makefile (now run with make instead of make run)
2018-07-16 16:31:59 -07:00
b5806d33db
Refactor RANDOM
2018-07-16 16:03:04 -07:00
044cd81ce6
Fix wrong conditions
2018-07-14 12:34:03 +02:00
1a61bdb9c0
Update zstd.lua
2018-07-14 12:27:42 +02:00
0e5fbc10fa
Update README
2018-07-13 17:41:09 -07:00
31731df4da
Remove clevel and update documentation
2018-07-13 17:38:53 -07:00
a23a3b95f9
Add random dictionary builder
2018-07-13 16:05:14 -07:00
23d77c531e
Added premake4/GENie script to contrib folder
2018-07-11 18:02:18 +02:00
87579d51ee
meson: fix build
2018-07-02 15:06:46 +01:00
bf30b9caf4
Add CXXFLAGS to ALL_LDFLAGS ( #1178 )
...
pzstd requires C++11, which older C++ standard libraries like libstdc++
as used on OS X 10.8 and earlier don't support. The user might address
this by setting "CXXFLAGS=-stdlib=libc++". This flag must be used both
at compile time and at link time. Asking the user to also put the flag
in LDFLAGS is undesirable because then the flag would also be used when
linking C code, which would be inappropriate.
2018-06-14 13:05:33 -07:00
750ee87a92
Merge pull request #1175 from ryandesign/macos
...
Fix name of macOS
2018-06-13 11:32:06 -04:00
8ebd0f3a17
Remove duplicate CXXFLAGS
2018-06-10 18:04:21 -05:00
b567ce9d68
Fix name of macOS
2018-06-09 14:31:17 -05:00
97c60cdf36
fixed seekable_format type mismatch
...
and some minor "unused variable" warnings.
Also : zstd_seekable.h is actually depending on zstd.h for ZSTDLIB_API
2018-06-06 13:10:29 -07:00
30ee23e905
ensure seekable_format/examples generated libzstd.a
...
when it's not already present in the expected directory
2018-06-06 12:09:58 -07:00
2dbe408a49
Make the build reproducible
...
Whilst working on the Reproducible Builds effort [0], we noticed
that zstd could not be built reproducibly.
This is due to the manual page encoding the number of CPUs from the
build machine and thus varies across builds.
This was originally filed in Debian as #897904 [1].
[0] https://reproducible-builds.org/
[1] https://bugs.debian.org/897904
Signed-off-by: Chris Lamb <lamby@debian.org>
2018-05-04 08:39:51 -07:00
00d52282d0
Use -pthread rather than -lpthread.
...
It can have other effects, such as pulling in extra libraries. Without it,
riscv build fails with undefined reference to `__atomic_compare_exchange_1'.
2018-04-09 01:50:49 +02:00
9e6ba88363
moved snap to /contrib
2018-03-19 16:15:06 -07:00
a7b532a347
added docker readme
2018-03-19 16:13:12 -07:00
b06db3b3c5
Merge pull request #1052 from Varunram/dockerfile
...
Add Dockerfile
2018-03-19 16:07:41 -07:00
7616200eaf
Add Dockerfile
...
Dockerfile initially proposed by @gyscos at #880
2018-03-18 14:53:48 +05:30
5373e44ba7
fixed contrib/adaptive-compression
2018-03-15 17:10:15 -07:00
355cb645bf
fixed seekable format example
2018-03-15 16:29:28 -07:00
38cbcb5f1a
removed LRM exploratory experiment
2018-03-15 16:26:08 -07:00
e916b9090e
gen_html: changed CFLAGS for CXXFLAGS
...
since it's associated with $(CXX)
2018-03-09 11:52:14 -08:00
ea41b580eb
meson: allow -Dlegacy_support=true, fix -Dlegacy_support=0
2017-12-29 10:18:33 -08:00
6ff43c0051
get soversion right
2017-12-24 10:05:43 -08:00
daffe435c0
meson: support differn't legacy levels.
...
Default to v0.4.0+
2017-12-21 15:47:38 -08:00
3ddfa42fe8
meson: fix build
...
used absolute paths which are deprecated in meson, also missing some sources
that got split
also move source files each to their own line so future diffs are clearer.
2017-12-19 22:02:03 -08:00
c173dbd6e7
no longer supported starting C++17
2017-12-04 18:00:53 -08:00
0d58aaf6f0
/contrib: fixed license header
...
removed last reference to PATENTS file
2017-10-02 02:07:17 -07:00
f1a3ce34b7
Merge pull request #826 from terrelln/license
...
[linux-kernel] Update license
2017-08-31 12:52:23 -07:00
f9252d8347
[linux-kernel] Update license
2017-08-31 12:48:36 -07:00
b0cb081dc8
last batch of header files changed to reflect new license ( #825 )
...
only remains to update contrib/linux-kernel (@terrelln)
2017-08-31 12:20:50 -07:00
e21384fffb
fixed more file headers after license change ( #825 )
2017-08-31 12:11:57 -07:00
e9dc204f42
fixed a bunch of headers after license change ( #825 )
2017-08-31 11:24:54 -07:00
394bdd7db9
changed license for examples
...
intentionnally this time
2017-08-29 09:24:11 -07:00
dd63533b29
Merge pull request #797 from paulcruz74/documentation
...
Updating README.md
2017-08-16 00:37:55 -07:00
733ca51360
Updating README.md
2017-08-15 17:48:23 -07:00
2daeab12f4
Merge pull request #795 from terrelln/squashfs-tools2
...
[kernel] Update squashfs-tools patch
2017-08-15 11:59:41 -07:00
57e2df6651
[kernel] Update squashfs-tools patch
2017-08-14 22:43:36 -07:00
fc90469587
updated program name print statement
2017-08-10 16:11:59 -07:00
75c3ad947c
Merge remote-tracking branch 'upstream/dev' into dev
2017-08-10 14:46:31 -07:00
8915b9b518
Merge pull request #786 from terrelln/squashfs-tools
...
[linux-kernel] Update patches for v4 and v5
2017-08-09 16:00:11 -07:00
8b6702a00d
[linux-kernel] Update patches for v5
2017-08-09 13:03:40 -07:00
a1fe656165
Merge pull request #779 from paulcruz74/adapt-approach-4
...
Patching warnings, adding ability to bound compression variation
2017-08-07 14:50:38 -07:00
7069bb9458
Merge branch 'adapt-approach-4' into bug-fixes
2017-08-07 13:18:53 -07:00
e100a311eb
removed direct assignment of 22, used ZSTD_maxCLevel() instead
2017-08-07 13:11:07 -07:00
7393b49fbd
[linux-kernel] Update patches for v4
2017-08-04 16:57:03 -07:00
01237e3b35
changed multi to zstd-adaptive in the help menu
2017-08-03 15:13:49 -07:00
8be7bba08c
added mutex for compression level to avoid data race
2017-08-02 10:27:33 -07:00
69ef22c0ac
added detach statements to prevent resource leak
2017-08-01 17:36:13 -07:00
f8c7b191e7
Merge branch 'adapt-approach-4' into dev
2017-07-31 15:42:57 -07:00
0295737ad7
change signal to broadcast for jobCompressed condition varaible since multiple threads waiting
2017-07-31 13:43:03 -07:00
cf92221671
Merge remote-tracking branch 'upstream/dev' into dev
2017-07-31 11:43:40 -07:00
c2f1bda8a1
Merge branch 'adapt-approach-4' into dev
2017-07-31 11:43:01 -07:00
9ea7df03de
add install target in makefile
2017-07-31 11:04:17 -07:00
f60cd3f99b
print defaults and range, remove EXT
2017-07-31 09:47:09 -07:00
e22b60cb76
removed ternary operation, added assert statement, check to make sure initial compression level is within bounds
2017-07-28 17:46:51 -07:00
cb9af53e77
delete empty line
2017-07-28 17:28:25 -07:00
51788225db
remove exe extension from makefile, reinclude pthread flag
2017-07-28 17:27:36 -07:00
ee27f6937c
Merge pull request #758 from stellamplau/ldm
...
Add working prototype of a long distance matcher
2017-07-28 17:07:21 -07:00
4d904ac800
add flags for multithreading
2017-07-28 16:12:58 -07:00
0f4cb67b00
add tests for compression bounds, fix another warning
2017-07-28 15:55:02 -07:00
ff54fced64
patched style errors, add ability to bound compression level variation
2017-07-28 15:30:46 -07:00
8fae41c412
Return error code in verify() and minor code cleanup
2017-07-27 17:14:05 -07:00
1294a4a897
Fix typo
2017-07-27 15:49:46 -07:00
627621839c
Add checks in initialization code
2017-07-27 15:37:37 -07:00
c105f605e6
Update README
2017-07-27 11:11:35 -07:00
e1222544be
Merge pull request #753 from paulcruz74/adapt-approach-3
...
adaptive compression v1
2017-07-27 10:00:10 -07:00
2320e7378a
remove unused variable, add documentation for context fields
2017-07-26 17:02:47 -07:00
9eaf3d22d0
Allow HASH_ONLY_EVERY_LOG to be configured in ldm.h
2017-07-26 16:43:25 -07:00
ab5a78547e
fix leaky abstraction regarding measuring completion
2017-07-26 16:40:05 -07:00
715f36ca81
added definitions for conversion constants, moved forced compression check to top of adaptCompressionLevel, used ZSTD_BLOCKSIZE_MAX
2017-07-26 15:52:15 -07:00
6c1c1242fc
set the window log value before performing compression
2017-07-26 14:29:59 -07:00
40759bade9
Add README and clean up code
2017-07-26 13:24:03 -07:00
a959cc881a
moved reset of completion to right after wait
2017-07-26 10:34:48 -07:00
305d5ee70f
change to >= convergence counter
2017-07-26 10:20:29 -07:00
be92a38d6a
decrease completion requirements for change, move create thread wait, merge cases where compression thread should wait
2017-07-26 10:05:10 -07:00
e9161637b2
Allow parameters to be modified from a separate file
2017-07-25 18:13:27 -07:00
0b18d21e03
building on readme, added another help tip in the menu
2017-07-25 17:47:02 -07:00
7cc74e0b27
adding more to readme
2017-07-25 16:55:16 -07:00
8dbb07d822
updated progress bar with better representation of time, added const
2017-07-25 16:03:43 -07:00
9a132707af
changing time units to seconds
2017-07-25 15:26:26 -07:00
629c300118
Rename and remove unneeded files
2017-07-25 15:17:36 -07:00
31a9ed9883
updated const values, added more comments
2017-07-25 14:53:40 -07:00
5cfbf609a4
removed old debug statements no longer being used
2017-07-25 14:31:48 -07:00
0882cd1981
progress bar -- don't print num jobs, time elapsed shown in seconds
2017-07-25 14:26:55 -07:00
310c12d07e
moved debug statements to a compiler flag
2017-07-25 14:08:39 -07:00
e02c79f833
started using decrease cooldown so that compression level would not decrease several times in a row
2017-07-25 11:16:27 -07:00
85d7c919f6
created independent function for controlling how completion relates to compression level change
2017-07-25 10:32:14 -07:00
6f1e260edd
added mechanism for getting rid of spikes
2017-07-25 10:01:10 -07:00
700758d676
added help statement for -p, switched it to hide progress bar now that progress bar is default
2017-07-24 16:26:20 -07:00
df3754b6ed
add quiet option, make progress bar default
2017-07-24 16:19:07 -07:00
0295a27133
Experiment with not using a checksum
2017-07-24 15:26:44 -07:00
4dc83ca64c
compression thread should take measurements independently based on whether or not the create/write thread will actually bottleneck performance
2017-07-24 15:14:58 -07:00
0ee3f8c2f8
adding more debug
2017-07-24 15:06:11 -07:00
8328f8192a
updating debug statements again
2017-07-24 14:40:23 -07:00
d3d759301f
changing position of endline for debug
2017-07-24 13:47:39 -07:00
08a6e9a141
Minor code cleanup
2017-07-24 13:22:00 -07:00
6eefa32911
Deduplicate code
2017-07-24 12:40:59 -07:00
8ed9220102
Experiment with eviction policies and minor code cleanup
2017-07-24 12:05:43 -07:00
e508f632d6
updated comments and debug statements
2017-07-24 11:01:36 -07:00
eb16da647d
Minor clean up
2017-07-24 10:20:53 -07:00
1a188fe864
Fix overflow bug when calculating hash
2017-07-24 10:20:53 -07:00
0b8fb1703b
Experiment with 64-bit hash insertion policy
2017-07-24 10:20:30 -07:00
273c17b350
Experiment with 64-bit hash and checksum
2017-07-24 10:19:50 -07:00
483d936b87
reduced competition for completion mutex by separating mutex use based on which values is updated
2017-07-23 14:09:16 -07:00
880f08d104
change how completion is measured in compression thread
2017-07-23 10:18:54 -07:00
08d9e42ec6
removed useless measurements
2017-07-21 18:02:55 -07:00
95bef759b3
switched over to model where reading only waits on compression thread
2017-07-21 17:49:39 -07:00
6455ec482c
taking the maximum of the completion level reads in order to determine which one was waiting more
2017-07-21 16:05:01 -07:00
05fe8dd47c
updating debug statements
2017-07-21 14:06:24 -07:00
db109f8fef
measure multiple completion levels during each wait
2017-07-21 13:38:24 -07:00
721c6a8b97
added bounding to compression level change
2017-07-21 09:30:24 -07:00
e929d3b787
added priority decision making for adapt compression level
2017-07-21 09:26:35 -07:00
9259c7afa4
semi working version that stabilizes
2017-07-20 18:45:33 -07:00
82e488770c
fixed bug where writeSize could be zero
2017-07-20 16:38:02 -07:00
a19916425d
reworked adaptCompressionLevel to only account for completion information
2017-07-20 16:19:16 -07:00
7d3ac0710d
[linux] Update patches for v3
2017-07-20 13:33:55 -07:00
7ab758a640
changed how completion is actually sampled
2017-07-20 10:53:51 -07:00
13a01ffb27
Fix off-by-one in size calculations
2017-07-19 17:24:09 -07:00
2427a154cb
Minor refactoring
2017-07-19 16:56:28 -07:00
dcf609f835
make adaptCompressionLevel oscillate less
2017-07-19 16:36:33 -07:00
2a22c7915e
call ZSTD_compressBegin() once
2017-07-19 16:00:54 -07:00
6767abe652
fixing error when file size is multiple of job size (in which case, the srcSize of the last job is 0)
2017-07-19 14:54:15 -07:00
030264ca51
Experiment with integrating ZSTD_count with findBestMatch
2017-07-19 14:14:26 -07:00
Lzu Tao
Yann Collet
Rohit Jain
Yann Collet
Nick Terrell
ko-zu
Azat Khuzhin
Jennifer Liu
Jennifer Liu
Codecat
Jon Turney
Ryan Schmidt
Ryan Schmidt
Chris Lamb
Adam Borowski
Varunram
Shawn Landden
Nick Terrell
Paul Cruz
Paul Cruz
Stella Lau