Commit Graph

4866 Commits

Author SHA1 Message Date
Yann Collet
c72091556b fixed minor nit as per @terrelln's comments 2018-02-09 09:46:08 -08:00
Yann Collet
4beaeaace5 Merge branch 'dev' into flexibleLevel 2018-02-09 09:15:05 -08:00
Yann Collet
6bfe50ad48 re-enabled ZSTD_decompressSequencesLong() 2018-02-09 09:14:25 -08:00
Yann Collet
1850597eaa pre-calculated default decoding tables 2018-02-09 06:01:02 -08:00
Yann Collet
ab75df21ed fixed mono-symbol distribution 2018-02-09 05:12:13 -08:00
Yann Collet
421a2716d8 fixed default fse distributions
but would be better to pre-calculate tables, for speed
2018-02-09 04:50:58 -08:00
Yann Collet
95424409ea addBits and baseline into FSE decoding table
note : unfinished
- need new default tables
- need modify long mode
2018-02-09 04:25:15 -08:00
Yann Collet
cc61a3694a Merge branch 'dev' into fasterDec 2018-02-09 02:41:02 -08:00
Yann Collet
d6e841d609 fixed streaming_memory_usage example
also:
ensure zstd.h is read from ../lib (instead of /usr/include)
2018-02-07 23:42:09 -08:00
Yann Collet
de68c2ff10 Merged ZSTD_preserveUnsortedMark() into ZSTD_reduceIndex()
as it's faster, due to one memory scan instead of two
(confirmed by microbenchmark).

Note : as ZSTD_reduceIndex() is rarely invoked,
it does not translate into a visible gain.
Consider it an exercise in auto-vectorization and micro-benchmarking.
2018-02-07 14:22:35 -08:00
Yann Collet
0170cf9a7a minor : modified ZSTD_preserveUnsortedMark() to be more vectorization friendly 2018-02-05 11:46:02 -08:00
Yann Collet
94efb1749d faster decoding in 32-bits mode for long offsets (tentative)
On my laptop:
Before:
./zstd32 -b --zstd=wlog=27 silesia.tar enwik8 -S
 3#silesia.tar       : 211984896 ->  66683478 (3.179),  97.6 MB/s , 400.7 MB/s
 3#enwik8            : 100000000 ->  35643153 (2.806),  76.5 MB/s , 303.2 MB/s

After:
./zstd32 -b --zstd=wlog=27 silesia.tar enwik8 -S
 3#silesia.tar       : 211984896 ->  66683478 (3.179),  97.4 MB/s , 435.0 MB/s
 3#enwik8            : 100000000 ->  35643153 (2.806),  76.2 MB/s , 338.1 MB/s

Mileage vary, depending on file, and cpu type.
But a generic rule is : x86 benefits less from "long-offset mode" than x64,
maybe due to register pressure.
On "entropy", long-mode is _never_ a win for x86.
On my laptop though, it may, depending on file and compression level
(enwik8 benefits more from "long-mode" than silesia).
2018-02-04 01:49:31 -08:00
Yann Collet
5188749e1c ensure compression parameters are updated when only compression level is changed 2018-02-02 16:31:20 -08:00
Yann Collet
4b525af53a zstdmt: applies new parameters on the fly
when invoked from ZSTD_compress_generic()
2018-02-02 15:58:13 -08:00
Yann Collet
90eca318a7 fileio: create dedicated function to generate zstd frames
like other formats
2018-02-02 14:24:56 -08:00
Yann Collet
1291d9d7cf
Merge pull request #1006 from systemcrash/patch-2
Update README.md
2018-02-02 10:04:55 -08:00
Yann Collet
549d26ae71
Merge pull request #1005 from systemcrash/dev
Update zstd.1
2018-02-02 10:04:40 -08:00
Yann Collet
6c492af284 fixed minor conversion warning 2018-02-01 20:16:00 -08:00
Yann Collet
209df52ba2 Changed nbThreads for nbWorkers
This makes it easier to explain that nbWorkers=0 --> single-threaded mode,
while nbWorkers=1 --> asynchronous mode (one mode thread on top of the "main" caller thread).
No need for an additional asynchronous mode flag.
nbWorkers>=2 works the same as nbThreads>=2 previously.
2018-02-01 19:29:30 -08:00
Yann Collet
4b6a94f0cc clarified comments on LDM parameters 2018-02-01 17:07:27 -08:00
Yann Collet
60fa90b6c0 zstdmt: added ability to change compression parameters during compression 2018-02-01 16:13:31 -08:00
Yann Collet
2bfc79ab8d removed bitstream.h dependency 2018-02-01 16:13:04 -08:00
Yann Collet
f25351e23e
Merge pull request #1009 from terrelln/fail-early
Test for incorrect pledgeSrcSize earlier
2018-02-01 14:43:32 -08:00
Nick Terrell
48acaddff9 Test for incorrect pledgeSrcSize earlier 2018-02-01 12:04:05 -08:00
Yann Collet
727bb7f090
Merge pull request #1008 from terrelln/hlog3
Fix hashLog3 size when copying cdict tables
2018-01-31 12:49:07 -08:00
Nick Terrell
9c8c69e41a [fuzzer] Check ZSTD_initStaticCDict() for every level 2018-01-31 11:12:54 -08:00
Nick Terrell
ab3346af07 Fix hashLog3 size when copying cdict tables 2018-01-31 11:12:17 -08:00
Yann Collet
823a28a1f4
Merge pull request #1000 from facebook/progressiveFlush
Progressive flush
2018-01-30 22:49:47 -08:00
Yann Collet
b173b8d8da updated zstd api manual 2018-01-30 15:05:12 -08:00
Yann Collet
a2ba629971 fixed function declaration ZSTD_getBlockSize() 2018-01-30 15:03:39 -08:00
Yann Collet
2cb0740b6b zstdmt: changed naming convention
to avoid confusion with blocks.

also:
- jobs are cut into chunks of 512KB now, to reduce nb of mutex calls.
- fix function declaration ZSTD_getBlockSizeMax()
- fix outdated comment
2018-01-30 14:43:36 -08:00
systemcrash
6b57387728
Update README.md
spelling
2018-01-29 18:42:20 +01:00
systemcrash
d13a75c969
Update zstd.1 2018-01-29 18:38:02 +01:00
Yann Collet
9f8ed23b5b bumped version number to v1.3.4
also added a paragraph on using compression level with training mode
as this is a recurrent question (see for example #1004)
2018-01-27 22:23:26 -08:00
Yann Collet
ba0cd8cf78 fixed minor conversion warning for C++ compilation mode 2018-01-26 18:18:42 -08:00
Yann Collet
caf9e96dc3 job mutex creation is checked 2018-01-26 18:09:25 -08:00
Yann Collet
9c40ae7ff1 zstdmt: there is now one mutex/cond per job 2018-01-26 17:55:08 -08:00
Yann Collet
77e36273de zstdmt: minor code refactor for clarity 2018-01-26 17:08:58 -08:00
Yann Collet
27c5853c42 zstdmt: job table correctly cleaned after synchronous ZSTDMT_compress() 2018-01-26 14:35:54 -08:00
Yann Collet
0d426f6b83 zstdmt : refactor a few member names
for clarity
2018-01-26 13:00:14 -08:00
Yann Collet
79b6e28b0a zstdmt : flush() only lock to read shared job members
Other job members are accessed directly.
This avoids a full job copy, which would access everything,
including a few members that are supposed to be used by worker only,
uselessly requiring additional locks to avoid race conditions.
2018-01-26 12:15:43 -08:00
Yann Collet
d2b62b6fa5 minor : ZSTDMT_writeLastEmptyBlock() is a void function
because it cannot fail
2018-01-26 11:06:34 -08:00
Yann Collet
fca13c6855 zstdmt : fixed memory leak
writeLastEmptyBlock() must release srcBuffer
as mtctx assumes it's done by job worker.

minor : changed 2 job member names (src->srcBuffer, srcStart->prefixStart) for clarity
2018-01-26 10:44:09 -08:00
Yann Collet
8e128eaf05 zstdmt : refactor job members
grouped by sharing properties
2018-01-26 10:20:38 -08:00
Yann Collet
777d3c1559 fixed minor declaration-after-statement warning 2018-01-25 17:45:18 -08:00
Yann Collet
a1d4041e69 zstdmt: removed job->jobCompleted
replaced by equivalent signal job->consumer == job->srcSize.

created additional functions
ZSTD_writeLastEmptyBlock()
and
ZSTDMT_writeLastEmptyBlock()
required when it's necessary to finish a frame with a last empty job, to create an "end of frame" marker.

It avoids creating a job with srcSize==0.
2018-01-25 17:35:49 -08:00
Yann Collet
1272d8e760 zstdmt:: renamed mutex and cond to underline they are context-global 2018-01-25 14:52:34 -08:00
Yann Collet
0fa96a6b78
Merge pull request #1001 from ne-sted/fix-defaults-align
cli: fix align of defaults
2018-01-24 09:34:36 -08:00
ne-sted
50aea2f293 cli: fix align of defaults 2018-01-24 15:07:22 +02:00
Yann Collet
4f7c896113 zstdmt : fixed complex sequencing bug
zstdmt would shortcut to single-thread blocking mode
in some rare cases where data is sent to be compressed but is not yet ready.
2018-01-23 18:00:51 -08:00