Commit Graph

1981 Commits

Author SHA1 Message Date
Nick Terrell
bbe77212ef [libzstd] Increase MaxOff 2017-09-25 13:36:18 -07:00
Yann Collet
96f0cde31a minor function rename
ZSTD_estimateCStreamSize_advanced_usingCParams -> ZSTD_estimateCStreamSize_usingCParams
_usingX is clear.
_advanced feels redundant
2017-09-24 16:47:02 -07:00
Yann Collet
7c3dea42ce added prototypes for advanced parameters for decompression API
required to decode custom formats
2017-09-24 15:57:29 -07:00
Yann Collet
e60f48c549 Merge branch 'dev' into newFormats 2017-09-24 14:33:37 -07:00
Yann Collet
8977224b9b Merge pull request #859 from terrelln/31
Prepare for ZSTD_WINDOWLOG_MAX == 31
2017-09-22 09:01:39 -07:00
Nick Terrell
d6abb28951 Prepare for ZSTD_WINDOWLOG_MAX == 31 2017-09-21 17:18:41 -07:00
Yann Collet
cd3115b284 added control from frame content size at end of decompression
adding check at end of single-pass ZSTD_decompressFrame().
Check within ZSTD_decompressContinue() was already added in a previous patch : b3f33ccfb3
2017-09-21 16:21:10 -07:00
Yann Collet
645563583e Merge branch 'dev' into newFormats 2017-09-21 16:08:06 -07:00
Yann Collet
f97c2dbd39 created ZSTD_format declaration 2017-09-21 16:07:29 -07:00
Yann Collet
da74aabc00 Merge pull request #850 from terrelln/fse-optimal
[fse] Fix FSE_optimalTableLog() for srcSize==1
2017-09-19 14:59:21 -07:00
Yann Collet
c399ab4804 Merge pull request #849 from terrelln/30
[bitstream] Allow adding 31 bits at a time
2017-09-19 14:25:10 -07:00
Nick Terrell
74718d7e43 [bitstream] Allow adding 31 bits at a time 2017-09-19 13:57:33 -07:00
Nick Terrell
6c9ed76676 [ldm] Fix corner case where minMatch < 8
There is a potential read buffer overflow when minMatch < 8.

fix-fuzz-failure
2017-09-19 13:49:37 -07:00
Nick Terrell
18442a31ff [libzstd] Fix bad window size assert
The window size is not validated or used in the one-pass API, so there
shouldn't be an assert based on it.

fix-fuzz-failure
2017-09-19 13:47:59 -07:00
Yann Collet
cb8b471e8b Merge branch 'dev' of github.com:facebook/zstd into dev 2017-09-18 14:48:23 -07:00
Yann Collet
7d1ff3817b fix ZSTD_sizeof_CCtx() / ZSTD_sizeof_CStream()
previous result was over-estimated
by counting streaming buffers twice
2017-09-18 14:47:34 -07:00
Nick Terrell
cae3e3c652 [fse] Fix FSE_optimalTableLog() for srcSize==1 2017-09-18 14:11:18 -07:00
Yann Collet
72a80515ec Merge pull request #848 from terrelln/fparams
[block] Don't use fParams in ZSTD_decompressBlock()
2017-09-18 13:48:31 -07:00
Yann Collet
539b91ee9b minor : added assert in bt 2017-09-16 23:41:58 -07:00
Nick Terrell
5f22479517 [block] Don't use fParams in ZSTD_decompressBlock() 2017-09-15 17:37:20 -07:00
Yann Collet
77c137b3ae minor comment refactor 2017-09-14 15:12:57 -07:00
Yann Collet
335780c427 fixed too strong alignment assert in ZSTD_initStaticCCtx()
64-bits fields are only 32-bits aligned on 32-bits CPU
2017-09-13 16:35:29 -07:00
Yann Collet
f1571dad8f Merge pull request #838 from stellamplau/ldm-mergeDev
Add long distance matcher
2017-09-13 13:24:08 -07:00
Yann Collet
4120a7fd5a Merge pull request #837 from facebook/libzstd-nomt
makes it possible to compile libzstd in single-thread mode without zs…
2017-09-12 17:13:17 -07:00
Yann Collet
3306bcb0e6 fix #820 : GCC v3.x 32-bits doesn't define 64-bits intrinsic
resulting in undefined symbol error.
Push the requirement to GCC 4 for now.
Another solution, proposed by @NWilson, is to use __LONG_MAX__ instead.
__LONG_MAX__ is a GCC-specific constant, which value is supposed to depend on underlying target hardware (32/64 bits)
Might be better, but seems also more complex, hence more prone to side effects.
Keeping the simple solution for now (just rely on __GNUC__)
2017-09-11 15:17:31 -07:00
Stella Lau
eb3327c10a Merge branch 'dev' of https://github.com/facebook/zstd into ldm-mergeDev 2017-09-11 15:00:01 -07:00
Stella Lau
f902bf9676 Merge branch 'ldm-integrate' into ldm-mergeDev 2017-09-11 14:55:29 -07:00
Yann Collet
f325ee4e84 fixed pass-through warning 2017-09-11 14:37:03 -07:00
Stella Lau
0d1b54db61 Explicitly cast raw numerals when left-shifting 2017-09-11 14:28:18 -07:00
Yann Collet
0d6ecc72a3 makes it possible to compile libzstd in single-thread mode without zstdmt_compress.c (#819) 2017-09-11 14:09:34 -07:00
Yann Collet
ce31004f20 fix following suggestions by @terrelln 2017-09-11 13:12:52 -07:00
Yann Collet
b3f33ccfb3 use ZSTD_decodingBufferSize_min() inside ZSTD_decompressStream()
Use same definition as public one
minor : reduce allocated buffer size in some cases
(when frameContentSize is known and == windowSize)
2017-09-09 14:37:28 -07:00
Yann Collet
058ed2ad33 ZSTD_decodingBufferSize_min()
supporting function for bufferless streaming API (ZSTD_decompressContinue())
makes it possible to correctly size a round buffer for decoding using this API.

also : added field blockSizeMax within ZSTD_frameHeader,
as it's a necessary information to know when to restart at beginning of decoding buffer.
2017-09-09 01:03:29 -07:00
Yann Collet
3128e03be6 updated license header
to clarify dual-license meaning as "or"
2017-09-08 00:09:23 -07:00
Stella Lau
360428c5d9 Move ldm functions to their own file 2017-09-06 18:09:26 -07:00
Yann Collet
baa37c3362 programs/Makefile : better support for GNU conventions
see https://www.gnu.org/prep/standards/html_node/Command-Variables.html
2017-09-06 16:53:59 -07:00
Yann Collet
3a12531a3d lib/Makefile : better support for GNU conventions
see https://www.gnu.org/prep/standards/html_node/Makefile-Conventions.html
2017-09-06 16:35:49 -07:00
Yann Collet
1c7b914cdf update README on BUCK file 2017-09-06 16:23:39 -07:00
Yann Collet
36374cc3b4 update and clarify lib/README 2017-09-06 16:15:18 -07:00
Stella Lau
2b99d696de Remove debug code 2017-09-06 15:57:26 -07:00
Stella Lau
eeff55dfa8 Merge remote-tracking branch 'upstream/dev' into ldm-mergeDev 2017-09-06 15:56:32 -07:00
Yann Collet
ad0046244f Merge pull request #831 from terrelln/split-compress
Split parsers out of zstd_compress.c
2017-09-06 10:01:27 -07:00
Stella Lau
9e4060200b Add tests and fix pointer alignment 2017-09-06 09:14:05 -07:00
Stella Lau
c706de5395 Rename and add short ldm parameters in cli 2017-09-05 21:11:18 -07:00
Stella Lau
98b85426f1 Fix setting of nextToUpdate at end of ldm matcher 2017-09-05 20:41:37 -07:00
Nick Terrell
721726d688 Split parsers out of zstd_compress.c 2017-09-05 17:10:25 -07:00
Stella Lau
08d33fe1c9 Fix parameter handling in copyCCtx with cdict 2017-09-05 15:50:20 -07:00
Stella Lau
fd0071da29 Fix parameter handling with ZSTD_copyCCtx 2017-09-05 15:34:17 -07:00
Stella Lau
643d28c701 Add ldm options to 'man zstd' 2017-09-05 11:27:15 -07:00
Nick Terrell
423b133568 [POOL] Allow free on NULL when multithreading is disabled 2017-09-05 11:18:13 -07:00
Stella Lau
67d4a6161c Add ldmBucketSizeLog param 2017-09-02 21:55:29 -07:00
Stella Lau
a1f04d518d Move hashEveryLog to cctxParams and update cli 2017-09-01 15:05:47 -07:00
Stella Lau
767a0b3be1 Move ldm hashLog, bucketLog, and mml to cctxParams 2017-09-01 12:24:59 -07:00
Yann Collet
8a5c0c98ae restored 32-bits decoder ability to decode long offsets (>32 MB, levels 21+) 2017-09-01 11:56:57 -07:00
Yann Collet
36aa8b5999 improved decoding speed 2017-09-01 11:40:59 -07:00
Stella Lau
17d8e0bdcc Merge remote-tracking branch 'upstream/longRangeMatcher' into ldm-integrate 2017-09-01 10:19:38 -07:00
Stella Lau
8081becadc Add long distance matching as a CCtxParam 2017-09-01 09:18:58 -07:00
Yann Collet
d963daa6a9 fixed minor warning (empty translation unit) 2017-09-01 00:12:07 -07:00
Yann Collet
3704507774 fixed decompression bug reported by @Etsukata (#828) 2017-09-01 00:05:37 -07:00
Yann Collet
369c29dd1a fixed impact of merge conflict for longRange 2017-08-31 18:25:56 -07:00
Yann Collet
d7ad99b2ab Merge branch 'longRangeMatcher' into dev 2017-08-31 18:08:37 -07:00
Stella Lau
6a546efb8c Add long distance matcher
Move last literals section to ZSTD_block_internal
2017-08-31 12:53:19 -07:00
Yann Collet
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
Yann Collet
e21384fffb fixed more file headers after license change (#825) 2017-08-31 12:11:57 -07:00
Yann Collet
e9dc204f42 fixed a bunch of headers after license change (#825) 2017-08-31 11:24:54 -07:00
Stella Lau
90a31bfa16 Pass dictMode to ZSTDMT_initCStream; fix nits
- Return error code in estimate{CCtx,CStream}Size functions
2017-08-30 16:19:07 -07:00
Stella Lau
ee65701720 Minor fixes; remove formatting only changes 2017-08-29 20:27:35 -07:00
Stella Lau
a6e20e1bd7 Add test for raw content starting with dict header 2017-08-29 18:36:18 -07:00
Stella Lau
623e3cd40b Use ZSTD_dm_rawContent in zstdmt_compress 2017-08-29 18:04:32 -07:00
Stella Lau
82d636b76a Rename applyCCtxParams() 2017-08-29 18:03:06 -07:00
Stella Lau
4e835720bf Delay creation of ZSTDMT_CCtx 2017-08-29 17:58:32 -07:00
Stella Lau
c7a18b7c21 Localize 'dictMode' from cctx to function param 2017-08-29 15:52:24 -07:00
Yann Collet
d6ddb879da Merge pull request #817 from terrelln/pool-custom-alloc
[pool] Accept custom allocators
2017-08-29 13:05:39 -07:00
Stella Lau
c88fb9267f Replace 'byReference' with enum 2017-08-29 11:55:02 -07:00
Nick Terrell
9822f97721 [error] Don't guard undef X with ifdef X 2017-08-29 11:54:38 -07:00
Stella Lau
b5b9275e67 Rename estimateCCtxSize_advanced() and estimateCStreamSize_advanced() 2017-08-29 10:49:29 -07:00
Stella Lau
0e56a84a1e Fix getting cParams from CCtxParams 2017-08-28 19:25:17 -07:00
Nick Terrell
02033be08c [pool] Visual Studios disallows empty structs 2017-08-28 17:19:01 -07:00
Nick Terrell
7c365eb02c [threading] Fix ERROR macro after including windows.h 2017-08-28 16:25:02 -07:00
Bernhard M. Wiedemann
cf689b84f9 Sort input file list
in order to make builds reproducible
in spite of indeterministic filesystem readdir order.
See https://reproducible-builds.org/ for why this is good.
2017-08-26 17:08:00 +02:00
Stella Lau
024098a47d Fix parameter retrieval from cdict 2017-08-25 17:58:28 -07:00
Stella Lau
2adde898c8 Fix typo with ZSTDMT_parameter 2017-08-25 16:13:40 -07:00
Stella Lau
18224608ff Remove ZSTD_setCCtxParameter() 2017-08-25 13:58:41 -07:00
Stella Lau
0744592d38 Add function initializing cctxParams from clevel 2017-08-25 13:36:47 -07:00
Stella Lau
9911153723 Move jobSize and overlapLog in zstdmt to cctxParams 2017-08-25 13:14:51 -07:00
Stella Lau
de5193422d Distinguish between jobParams and cctxParams in zstdmt 2017-08-25 11:36:17 -07:00
Stella Lau
eb7bbab36a Remove ZSTD_p_refDictContent and dictContentByRef 2017-08-25 11:11:45 -07:00
Nick Terrell
db3f5372df [zstdmt] Use POOL_create_advanced() 2017-08-24 18:12:28 -07:00
Nick Terrell
de6c6bce85 Fix zstd_internal.h for C++ mode 2017-08-24 18:09:50 -07:00
Nick Terrell
26dc040a7b [pool] Accept custom allocators 2017-08-24 17:01:41 -07:00
Nick Terrell
89dc856cae [pool] Fix formatting 2017-08-24 16:48:32 -07:00
Stella Lau
15fdeb9e41 Enforce nbThreads<=1 for estimateCCtxSize 2017-08-24 16:28:49 -07:00
Nick Terrell
376f435914 [dictBuilder] Set default compression level to 3 2017-08-24 16:21:05 -07:00
Stella Lau
2fbf0285b2 Fix interaction with ZSTD_setCCtxParameter() and cleanup 2017-08-24 11:25:41 -07:00
Stella Lau
fd9bf42516 Fix forceWindow and dictMode setting for zstdmt jobs 2017-08-23 19:16:57 -07:00
Stella Lau
bf3108fb50 Ensure zstdmt uses 'job version' of cctx parameters 2017-08-23 17:03:31 -07:00
Stella Lau
1c81f725ff Remove duplicated testing code 2017-08-23 15:47:15 -07:00
Stella Lau
64ce49426b Fix cstream compression level 2017-08-23 12:30:47 -07:00
Stella Lau
5bc2c1e982 Add prototype support for customMem with cctxParams 2017-08-23 12:03:30 -07:00
Yann Collet
e9ce1208a1 Merge pull request #812 from facebook/longRangeFix
fixed extraordinary scenario where all fields use maximum nbBits
2017-08-23 11:35:28 -07:00
Yann Collet
74cde5a4d8 Merge pull request #813 from stellamplau/highbit32fix
Fix undefined behavior when srcSize==1
2017-08-23 11:31:06 -07:00
Stella Lau
6f1a21c7e9 Remove formatting-only changes 2017-08-23 10:24:19 -07:00
Dmitriy Titarenko
20f715d709 Fix displayLevel overflow 2017-08-23 15:56:15 +05:00
Stella Lau
11303778d0 Add function to make cctxParams from ZSTD_parameters 2017-08-22 14:53:13 -07:00
Yann Collet
bd9c8ca146 Merge pull request #811 from terrelln/segmentSize
[cover] Fix end condition for small dictionary
2017-08-22 14:36:30 -07:00
Stella Lau
23fc0e41fa Remove 'opaque' naming from internal functions 2017-08-22 14:24:47 -07:00
Stella Lau
8fd1636776 Remove unused functions 2017-08-22 13:33:58 -07:00
Yann Collet
6b2b6a9bd5 fixed extraordinary scenario where all fields use maximum possible nb of bits simultaneously
can only happen if windowLog>=27  (level 22 --ultra)
2017-08-22 12:09:21 -07:00
Stella Lau
e50ed1fa3a Fix undefined behavior when srcSize==1 2017-08-22 11:55:42 -07:00
Stella Lau
60e1bc617c Explicitly create a job cctxParam for multithreading 2017-08-21 15:39:37 -07:00
Stella Lau
5b956f4753 Comment out CCtx_param versions of CDict functions 2017-08-21 14:49:16 -07:00
Nick Terrell
29c2d9a4d0 [cover] Turn down notification for ZDICT subroutines 2017-08-21 14:28:31 -07:00
Nick Terrell
98de3f6847 [cover] Add dictionary size to compressed size 2017-08-21 14:23:17 -07:00
Yann Collet
78c3d16bf4 Merge pull request #809 from terrelln/dev
[cover] Fix divide by zero
2017-08-21 13:33:19 -07:00
Nick Terrell
9a54a315aa [cover] Convert score to U32 and check for zero 2017-08-21 13:30:07 -07:00
Stella Lau
fd8a25786e Check parameters are valid in initCCtxParams 2017-08-21 13:23:35 -07:00
Stella Lau
1c0dbe81b1 Add documentation for CCtx_params 2017-08-21 13:18:00 -07:00
Nick Terrell
d49eb40c03 [cover] Stop when segmentSize is less than d 2017-08-21 13:10:03 -07:00
Stella Lau
939f954285 Pass ZSTD_CCtx_params as const ptr when possible 2017-08-21 12:57:18 -07:00
Stella Lau
73c73bf16a Reduce code duplication in zstreamtest 2017-08-21 12:41:19 -07:00
Stella Lau
560b34f6d2 Return error code when initializing NULL cctxParams 2017-08-21 11:52:26 -07:00
Stella Lau
25be09c6b4 Set some parameters to zero before initializing cdict 2017-08-21 11:35:46 -07:00
Yann Collet
232d62b637 fixed a few headers that were too hastily copy/pasted during last license change 2017-08-21 11:24:32 -07:00
Nick Terrell
f306d400c0 [cover] Fix divide by zero 2017-08-21 11:12:11 -07:00
Stella Lau
502031ca10 Use cctxParam version of createCDict internally 2017-08-21 11:00:44 -07:00
Stella Lau
91b30dbe84 Remove test parameter 2017-08-21 10:09:06 -07:00
Stella Lau
f181f33bdf Disable tests and refactor 2017-08-21 01:59:08 -07:00
Stella Lau
023b24e6d4 Add cctx param tests 2017-08-20 22:55:07 -07:00
Yann Collet
7db552676e reduced pool queue to 0 to save memory
fixed : pool performance when jobs are fires fast and queueSize==0
2017-08-19 15:07:54 -07:00
Stella Lau
6cee6e07e5 Add internal createCDict function 2017-08-18 22:48:31 -07:00
Stella Lau
d775519296 Add cctxParam versions of internal functions 2017-08-18 17:37:58 -07:00
Yann Collet
32fb407c9d updated a bunch of headers
for the new license
2017-08-18 16:52:05 -07:00
Stella Lau
63b8c98531 Pass cctx parameters to MTCtx 2017-08-18 16:17:24 -07:00
Stella Lau
399ae013d4 Add function to apply cctx params 2017-08-18 13:01:55 -07:00
Stella Lau
81d89d82a6 Move nbThreads to cctx params 2017-08-18 12:08:57 -07:00
Stella Lau
2300c58a6f Move dictContentByRef to cctx params 2017-08-18 12:03:16 -07:00
Stella Lau
b6cb2ed8cb Move dictMode to cctxParams 2017-08-18 11:43:31 -07:00
Stella Lau
97e27affcb Move compression level to cctx params 2017-08-18 11:20:08 -07:00
Stella Lau
c0221124d5 Add function to set opaque parameters 2017-08-17 19:30:22 -07:00
Stella Lau
4169f49171 Add initialization/allocation functions for opaque params 2017-08-17 18:45:04 -07:00
Stella Lau
ade95b8bed Add opaque interfaces for static initialization 2017-08-17 18:13:08 -07:00
Stella Lau
699f11b4f7 Create opaque parameter structure 2017-08-17 17:33:46 -07:00
Yann Collet
f9e6590715 Merge pull request #796 from terrelln/is-error
[FSE][HUF] Inline error checks
2017-08-15 12:37:28 -07:00
Yann Collet
2dbcfc6994 Merge pull request #794 from terrelln/force-inline
[libzstd] Fix FORCE_INLINE macro
2017-08-15 12:03:44 -07:00
Nick Terrell
07c6ff588e [FSE][HUF] Inline error checks
Caught by Clang's optimization remarks.
2017-08-15 11:23:28 -07:00
Nick Terrell
565e925eb7 [libzstd] Fix FORCE_INLINE macro 2017-08-14 21:12:05 -07:00
Roman Gershman
b9d4f4fb74 Fix ZSTD_estimateDStreamSize function after ZSTD_DStream and ZSTD_DCtx were merged 2017-08-13 13:29:42 +03:00
Nick Terrell
9ba97182d1 [CI] Add gcc7build test 2017-08-08 13:28:56 -07:00
Yann Collet
d9f2893eb9 Merge pull request #782 from terrelln/dstSizeTooSmall
Fix compression failure on incompressible data
2017-08-07 14:52:02 -07:00
Yann Collet
8049556928 Merge pull request #778 from terrelln/bad-huff
[libzstd] Fix bug in Huffman decompresser
2017-08-07 14:05:58 -07:00
Nick Terrell
abe12b3399 [libzstd] Fix bug in Huffman decompresser
The zstd format specification doesn't enforce that Huffman compressed
literals (including the table) have to be smaller than the uncompressed
literals. The compressor will never Huffman compress literals if the
compressed size is larger than the uncompressed size. The decompresser
doesn't accept Huffman compressed literals with 4 streams whose compressed
size is at least as large as the uncompressed size.

* Make the decompresser accept Huffman compressed literals whose size
  increases.
* Add a test case that exposes the bug. The compressed file has to be
  statically generated, since the compressor won't normally produce files
  that expose the bug.
2017-08-07 12:37:48 -07:00
Nick Terrell
308047eb5d Fix compression failure on incompressible data
If the destination buffer is the minimum allowed size in
`ZSTD_compressSequences()` (2^17), then if the block isn't compressible
compression might fail with `dstSize_tooSmall`, when it should instead emit
a raw uncompressed block.

Additionally, `ZSTD_compressLiterals()` implicitly called
`ZSTD_noCompressLiterals()` if Huffman compression failed. Make that
explicit.
2017-08-07 11:45:24 -07:00
Stella Lau
73ba58955f Signal after finishing job when queueSize=0 2017-08-01 20:12:06 -07:00
Stella Lau
1d76da1d87 Replace marker with queueEmpty variable and update pool.h comment 2017-08-01 12:30:16 -07:00
Stella Lau
5adceeed01 Allow queueSize=0 in pool.c and update poolTests 2017-07-31 10:10:16 -07:00
Yann Collet
e1222544be Merge pull request #753 from paulcruz74/adapt-approach-3
adaptive compression v1
2017-07-27 10:00:10 -07:00
Nick Terrell
ae20d413da [libzstd] Fix CHECK_V_F macros 2017-07-25 12:52:01 -07:00
Yann Collet
a90b16e150 Visual blind fix 2 2017-07-20 15:57:55 -07:00
Yann Collet
b4d460f32c pool.c : blindfix for Visual warnings 2017-07-20 01:13:14 -07:00
Yann Collet
3974d2b38a blind fix for Windows Multithreading module
adds a fake 0 return value for mutex/cond init
2017-07-19 13:33:21 -07:00
Paul Cruz
6945b3c43d removed previous version of completion for compression 2017-07-19 11:51:50 -07:00
Yann Collet
b71363b967 check pthread_*_init() success condition 2017-07-19 01:05:40 -07:00
Nick Terrell
cc1522351f [libzstd] Fix bug in Huffman encoding
Summary:
Huffman encoding with a bad dictionary can encode worse than the
HUF_BLOCKBOUND(srcSize), since we don't filter out incompressible
input, and even if we did, the dictionaries Huffman table could be
ill suited to compressing actual data.

The fast optimization doesn't seem to improve compression speed,
even when I hard coded fast = 1, the speed didn't improve over hard coding
it to 0.

Benchmarks:
$ ./zstd.dev -b1e5
Benchmarking levels from 1 to 5
 1#Synthetic 50%     :  10000000 ->   3139163 (3.186), 524.8 MB/s ,1890.0 MB/s
 2#Synthetic 50%     :  10000000 ->   3115138 (3.210), 372.6 MB/s ,1830.2 MB/s
 3#Synthetic 50%     :  10000000 ->   3222672 (3.103), 223.3 MB/s ,1400.2 MB/s
 4#Synthetic 50%     :  10000000 ->   3276678 (3.052), 198.0 MB/s ,1280.1 MB/s
 5#Synthetic 50%     :  10000000 ->   3271570 (3.057), 107.8 MB/s ,1200.0 MB/s
$ ./zstd -b1e5
Benchmarking levels from 1 to 5
 1#Synthetic 50%     :  10000000 ->   3139163 (3.186), 524.8 MB/s ,1870.2 MB/s
 2#Synthetic 50%     :  10000000 ->   3115138 (3.210), 370.0 MB/s ,1810.3 MB/s
 3#Synthetic 50%     :  10000000 ->   3222672 (3.103), 223.3 MB/s ,1380.1 MB/s
 4#Synthetic 50%     :  10000000 ->   3276678 (3.052), 196.1 MB/s ,1270.0 MB/s
 5#Synthetic 50%     :  10000000 ->   3271570 (3.057), 106.8 MB/s ,1180.1 MB/s
$ ./zstd.dev -b1e5 ../silesia.tar
Benchmarking levels from 1 to 5
 1#silesia.tar       : 211988480 ->  73651685 (2.878), 429.7 MB/s ,1096.5 MB/s
 2#silesia.tar       : 211988480 ->  70158785 (3.022), 321.2 MB/s ,1029.1 MB/s
 3#silesia.tar       : 211988480 ->  66993813 (3.164), 243.7 MB/s , 981.4 MB/s
 4#silesia.tar       : 211988480 ->  66306481 (3.197), 226.7 MB/s , 972.4 MB/s
 5#silesia.tar       : 211988480 ->  64757852 (3.274), 150.3 MB/s , 963.6 MB/s
$ ./zstd -b1e5 ../silesia.tar
Benchmarking levels from 1 to 5
 1#silesia.tar       : 211988480 ->  73651685 (2.878), 429.7 MB/s ,1087.1 MB/s
 2#silesia.tar       : 211988480 ->  70158785 (3.022), 318.8 MB/s ,1029.1 MB/s
 3#silesia.tar       : 211988480 ->  66993813 (3.164), 246.5 MB/s , 981.4 MB/s
 4#silesia.tar       : 211988480 ->  66306481 (3.197), 229.2 MB/s , 972.4 MB/s
 5#silesia.tar       : 211988480 ->  64757852 (3.274), 149.3 MB/s , 963.6 MB/s

Test Plan:
I added a test case to the fuzzer which crashed with ASAN before the patch
and succeeded after.
2017-07-18 13:20:40 -07:00
Yann Collet
77d67fb167 Merge pull request #766 from terrelln/real-block-split
[libzstd] Pull optimal parser state out of seqStore_t
2017-07-18 08:26:24 -07:00
Yann Collet
14c83b05c7 Merge pull request #765 from terrelln/real-block-split
[libzstd] Remove ZSTD_CCtx* argument of ZSTD_compressSequences()
2017-07-17 19:25:55 -07:00
Nick Terrell
7a28b9e4a3 [libzstd] Pull optimal parser state out of seqStore_t 2017-07-17 15:29:11 -07:00
Yann Collet
3381bf4b84 Merge pull request #764 from terrelln/real-block-split
[libzstd] Refactor ZSTD_compressSequences()
2017-07-17 14:46:01 -07:00
Nick Terrell
e198230645 [libzstd] Remove ZSTD_CCtx* argument of ZSTD_compressSequences() 2017-07-17 12:27:24 -07:00
Nick Terrell
634f012420 [libzstd] Refactor ZSTD_compressSequences() 2017-07-17 11:36:11 -07:00
Paul Cruz
50ce4eaeb6 added error detection for pthread initialization, added compression completion measurement, fixed const values 2017-07-17 10:12:44 -07:00
Yann Collet
3b0cff3c33 fixed clang's -Wdocumentation 2017-07-13 18:58:30 -07:00
Yann Collet
2bd6440be0 pinned down error code enum values
Note : all error codes are changed by this new version,
but it's expected to be the last change for existing codes.

Codes are now grouped by category, and receive a manually attributed value.
The objective is to guarantee that
error code values will not change in the future
when introducing new codes.
Intentionnal empty spaces and ranges are defined
in order to keep room for potential new codes.
2017-07-13 17:12:16 -07:00
Nick Terrell
830ef4152a [libzstd] Increase granularity of FSECTable repeat mode 2017-07-13 12:45:39 -07:00
Yann Collet
d985319337 Merge pull request #759 from terrelln/real-block-split
[libzstd] Pull CTables into sub-structure
2017-07-13 10:24:19 -07:00
Yann Collet
3a60efd3a9 policy change : ZSTDMT automatically caps nbThreads to ZSTDMT_NBTHREADS_MAX (#760)
Previously, ZSTDMT would refuse to create the compressor.
Also : increased ZSTDMT_NBTHREADS_MAX to 256,
updated doc,
and added relevant test
2017-07-13 10:17:23 -07:00
Yann Collet
132e6efd76 switched ZSTDMT_compress_advanced() last argument to overlapLog
overlapRLog (== 9 - overlapLog) was a bit "strange"
as all other public entry points use overlapLog
2017-07-13 02:22:58 -07:00
Yann Collet
4e77f7761d clarified comment on ZSTD_p_contentSizeFlag 2017-07-13 02:09:07 -07:00
Nick Terrell
de0414b736 [libzstd] Pull CTables into sub-structure 2017-07-12 19:49:19 -07:00
Yann Collet
8ef666c325 slightly increased buffer pool, to cover normal "full load" scenarios
2 buffers per active worker
+ 1 buffer for input loading
+ 1 buffer for "next input" when submitting current one
+ 1 buffer stuck in queue
2017-07-12 14:23:34 -07:00
Yann Collet
052a95f77c fix : ZSTDMT_compress_advanced() correctly generates checksum
when params.fParams.checksumFlag==1.
This use case used to be impossible when only ZSTD_compress() was available
2017-07-11 17:18:26 -07:00
Yann Collet
2a62f48bf4 release input buffers from inside worker thread
buffers are released sooner, which makes them available faster for next job.
=> decreases total nb of buffers necessary
2017-07-11 15:56:40 -07:00
Yann Collet
57236184af buffer pool : all buffers have same size
to reduce memory fragmentation.
They can be used for in or out, interchangeably.
2017-07-11 15:17:25 -07:00
Yann Collet
34b2b95631 zstdmt : intermediate outBuffer allocated from within worker
reduces total amount of memory needed,
since jobs in queue do not have an outBuffer pre-reserved now
2017-07-11 14:59:10 -07:00
Yann Collet
16261e6951 buffer pool can be invoked from multiple threads 2017-07-11 14:14:07 -07:00
Yann Collet
ef0ff7fe7f zstdmt: removed margin for improved memory usage 2017-07-11 08:54:29 -07:00
Yann Collet
4616fad18b improved ZSTDMT_compress() memory usage
does not need the input buffer for streaming operations

also : reduced a few tests time length
2017-07-10 17:16:41 -07:00
Yann Collet
670b1fc547 optimized memory usage for ZSTDMT_compress()
Previously, each job would reserve a CCtx right before being posted.
The CCtx would be "part of the job description",
and only released when the job is completed (aka flushed).
For ZSTDMT_compress(), which creates all jobs first and only join at the end,
that meant one CCtx per job.
The nb of jobs used to be == nb of threads,
but since latest modification,
which reduces the size of jobs in order to spread the load of difficult areas,
it also increases the nb of jobs for large sources / small compression level.
This resulted in many more CCtx being created.

In this new version, CCtx are reserved within the worker thread.
It guaranteea there cannot be more CCtx reserved than workers (<= nb threads).

To do that, it required to make the CCtx Pool multi-threading-safe :
it can now be called from multiple threads in parallel.
2017-07-10 16:30:55 -07:00
Yann Collet
3510efb02d fix : custom allocator correctly propagated to child contexts 2017-07-10 14:21:40 -07:00
Yann Collet
88da8f1816 fix : propagate custom allocator to ZSTDMT though ZSTD_CCtx_setParameter()
also : compile fuzzer with MT enabled
2017-07-10 14:02:33 -07:00
Yann Collet
e32fb0c1fe added ZSTD_sizeof_CCtx() test 2017-07-10 12:29:57 -07:00
Yann Collet
40156a4967 bumped version nb to v1.3.1 2017-07-08 04:55:09 -07:00
Yann Collet
0f4fc6c20a fixed several conversion warnings 2017-07-07 17:13:12 -07:00
Yann Collet
9bde061a0b fixed minor Visual compilation limitation 2017-07-07 16:14:17 -07:00
Yann Collet
593d517ebf fixed minor cast warning 2017-07-07 16:09:47 -07:00
Yann Collet
ead4dd48f6 new field frameHeader.headerSize 2017-07-07 15:51:24 -07:00
Yann Collet
46396523c0 ZSTD_getFrameHeader : control of windowSize limits is delegated to caller
Extracting frame header is a separate operation.
It's now possible to get frame header, whatever the window size set in it.
2017-07-07 15:32:12 -07:00
Yann Collet
990449b89d new field : ZSTD_frameHeader.frameType
Makes frame type (zstd,skippable) detection more straighforward.
ZSTD_getFrameHeader set frameContentSize=ZSTD_CONTENTSIZE_UNKNOWN to mean "field not present"
2017-07-07 15:21:35 -07:00
Yann Collet
e622330a3b extended frameHeader.windowSize to unsigned long long 2017-07-07 14:19:01 -07:00
Yann Collet
f04deff4fc fixed #718, reported by @GregSlazinski, solution suggested by @mcmilk 2017-07-06 01:42:46 -07:00
Yann Collet
d75c0e71c4 minor code refactoring 2017-07-05 18:10:07 -07:00