facebook
/
zstd
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
2,180 Commits
2 Branches
0 Tags
28 MiB
Commit Graph

2180 Commits (ad3e94512c3f53850cbe200f222720a4edf7ba8f)

Author SHA1 Message Date
Przemyslaw Skibinski 86dc7223e9 generate_res.bat: automatic detection if windres.exe is available 2016-10-28 12:20:02 +02:00
Przemyslaw Skibinski 3575fcf28e updated build\VS2010\zstd\generate_res\verrsrc.h 2016-10-28 11:54:47 +02:00
Yann Collet 5f1cb45cff Merge pull request #431 from terrelln/perf-regr 
Fix ZSTD_execSequence() performance regression
 2016-10-27 20:45:52 -07:00
Nick Terrell 10bfd0c0d5 Fix ZSTD_execSequence() performance regression 
Commit ae1cb3b3d0 caused the regression.
It is an instruction alignment issue, because if it is `U64 i` instead
of `U32 i`, the regression returns.  This patch fixes the regression
in gcc, but only gets some of the clang performance back.

Benchmarks:
Run on `silesia.tar`.  I only show levels 1-5 because the performance
regression was uniform across all levels.  I did one run on levels
1-19 and it looked good.

| Build | Level | Before | While | After |
|-------|-------|-------:|------:|------:|
| gcc   |     1 |  931.4 | 904.4 | 932.8 |
| gcc   |     2 |  849.1 | 822.6 | 851.2 |
| gcc   |     3 |  815.6 | 790.6 | 818.9 |
| gcc   |     4 |  794.1 | 770.7 | 798.0 |
| gcc   |     5 |  785.7 | 760.7 | 788.8 |
| clang |     1 |  705.5 | 683.2 | 693.8 |
| clang |     2 |  670.0 | 649.2 | 660.7 |
| clang |     3 |  659.6 | 639.8 | 651.4 |
| clang |     4 |  652.5 | 634.7 | 645.9 |
| clang |     5 |  646.9 | 625.5 | 637.7 |
 2016-10-27 16:19:57 -07:00
Yann Collet ee5b725823 ZSTD_initCStream() optimization : do not allocate a CDict when no dictionary used 2016-10-27 14:20:55 -07:00
Yann Collet bfae317ae3 minor const added 2016-10-26 18:44:59 -07:00
Yann Collet cf08958d9b print one global summary (instead of one per file) 2016-10-26 18:37:29 -07:00
Yann Collet 82613d72e4 added example multiple_streaming_compression 2016-10-26 18:10:43 -07:00
Przemyslaw Skibinski 55b50b0a68 Merge remote-tracking branch 'refs/remotes/facebook/dev' into dev11 2016-10-26 09:39:35 +02:00
Nick Terrell ac4310d303 [pzstd] Fix test mode for streaming input 2016-10-26 00:09:39 -07:00
Nick Terrell eb7873a048 ZSTD_execSequence() accepts match in last 7 bytes 
The zstd reference compressor will not emit a match in the last 7
bytes of a block.  The decompressor will also not accept a match
in the last 7 bytes.  This patch makes the decompressor accept a
match in the last 7 bytes.
 2016-10-25 21:24:15 -07:00
Yann Collet 335ad5d4d4 added ZSTD_initDStream_usingDDict() . 
slightly optimized ZSTD_initDStream() when no dictionary .
fixed ZSTD_sizeof_CStream() .
 2016-10-25 17:47:02 -07:00
Yann Collet 9516234e67 first sketch for ZSTD_initCStream_usingCDict() 2016-10-25 16:19:52 -07:00
Yann Collet 62d9a7ddfd Merge pull request #429 from inikep/btopt2 
Btopt2
 2016-10-25 14:48:43 -07:00
Yann Collet 57cca28026 Merge pull request #428 from inikep/dev11 
windres
 2016-10-25 11:34:09 -07:00
Przemyslaw Skibinski 5c5f01f3da added ZSTD_btopt2 strategy 2016-10-25 12:25:07 +02:00
Przemyslaw Skibinski 347903cfb5 Windows resources moved to programs/windres/ 2016-10-25 10:34:41 +02:00
Przemyslaw Skibinski 5fc901d1ce images/ moved to doc/images/ 2016-10-25 10:05:20 +02:00
Yann Collet 7b5948cca7 Merge pull request #426 from terrelln/fixes 
Fix various {A, M}SAN bugs
 2016-10-24 23:42:26 -07:00
Yann Collet 37d130031d updated comments on context re-use 2016-10-24 17:22:12 -07:00
Nick Terrell b2c39a22b0 Fix compiler narrowing warning 2016-10-24 14:50:13 -07:00
Nick Terrell f698ad6deb Merge remote-tracking branch 'upstream/dev' into fixes 
* upstream/dev:
  added doc\zstd_manual.html
  added contrib\gen_html
  zstd_compression_format.md moved to doc/
  Fix small bug in ZSTD_execSequence()
  improved ZSTD_compressBlock_opt_extDict_generic
  protect ZSTD_decodeFrameHeader() from invalid usage, as suggested by @spaskob
  zstd_opt.h: small improvement in compression ratio
  improved dicitonary segment merge
  use implicit rules to compile zstd_decompress.c
  detect early impossible decompression scenario in legacy decoder v0.5
  no repeat mode in legacy v0.5
  fixed invalid invocation of dictionary in legacy decoder v0.5
  fix edge case
  fix command line interpretation
  fixed minor corner case
  zstd.h: added the Introduction section
  fixed clang 3.5 warnings
  zstd.h: updated comments
 2016-10-24 13:10:13 -07:00
Yann Collet 4239a207dd Merge pull request #425 from inikep/dev11 
Doc
 2016-10-24 11:11:40 -07:00
Nick Terrell f9c9af3c2e Reject dictionaries with incomplete entropy tables 
If a dictionary specifies that a symbol has probability zero in its
`matchLength`, `literalLength`, or `offset` FSE table, but the symbol
appears when compressing input, the compressor fails.

Ensure that dictionaries support all `matchLength`, and `literalLength`
codes.  They must also support all of the `offset` codes required to
represent every possible offset that can appear in the first block.
 2016-10-24 10:42:44 -07:00
Przemyslaw Skibinski 86d9424c81 added doc\zstd_manual.html 2016-10-24 16:07:53 +02:00
Przemyslaw Skibinski 984b66cd72 added contrib\gen_html 2016-10-24 15:59:51 +02:00
Przemyslaw Skibinski 3ee94a7600 zstd_compression_format.md moved to doc/ 2016-10-24 15:58:07 +02:00
Yann Collet 97611611a3 Merge pull request #423 from terrelln/exec-seq-patch 
Fix small bug in ZSTD_execSequence()
 2016-10-21 17:02:06 -07:00
Nick Terrell ae1cb3b3d0 Fix small bug in ZSTD_execSequence() 
`memmove(op, match, sequence.matchLength)` is not the desired behavior.
Overlap is allowed, and handled as if we did `*op++ = *match++`, which
is not how `memmove()` handles overlap.

Only triggered if both of the following conditions are met:
* The match spans extDict & currentPrefixSegment
* `oLitEnd <= oend_w < oLitEnd + length1 < oMatchEnd <= oend`.

These two conditions imply that the block is less than 15 bytes long.
This bug isn't triggered by the streaming API, because it allocates
enough space for the window size + the block size, so there cannot be
a match that is within 8 bytes of the end and overlaps with itself.
It cannot be triggered by the block decompression API because all of
the decompressed data is in the currentPrefixSegment.

Introduced by commit 7158584399
 2016-10-21 12:13:44 -07:00
Przemyslaw Skibinski 4732074a71 improved ZSTD_compressBlock_opt_extDict_generic 2016-10-21 11:19:00 +02:00
Yann Collet da3bd8b6de protect ZSTD_decodeFrameHeader() from invalid usage, as suggested by @spaskob 2016-10-20 20:11:00 -07:00
Przemyslaw Skibinski d365ae3497 zstd_opt.h: small improvement in compression ratio 2016-10-20 11:49:02 +02:00
Przemyslaw Skibinski 575ab00db7 Merge remote-tracking branch 'refs/remotes/facebook/dev' into dev11 2016-10-20 11:01:52 +02:00
Przemyslaw Skibinski dafd66c7ac Merge remote-tracking branch 'refs/remotes/origin/dev' into dev11 2016-10-20 10:54:39 +02:00
Nick Terrell d760529a05 Fix stack buffer overrun when weightTotal == 0 
If `weightTotal == 0`, then `BIT_highbit32(weightTotal)` is
undefined behavior in the case that it calls `__builtin_clz()`.
If `tableLog == HUF_TABLELOG_ABSOLUTEMAX` then we will access one
byte beyond the end of the buffer.
 2016-10-19 11:39:11 -07:00
Nick Terrell bb68062c59 Unitialized memory read in ZSTD_decodeSeqHeaders() 
Caused by two things:
1. Not checking that `ip` is in range except for the first byte.
2. `ZSTDv0{5,6}_decodeLiteralsBlock()` could return a value larger than `srcSize`.
 2016-10-18 16:41:33 -07:00
Yann Collet 52c1bf93fe improved dicitonary segment merge 2016-10-18 16:34:58 -07:00
Yann Collet a7a4690b0a use implicit rules to compile zstd_decompress.c 2016-10-18 16:01:03 -07:00
Nick Terrell 7b06ad7a05 Backport fix from commit 125d817 
This fixes a read of unitialized memory.
Full commit hash: 125d81774f.
 2016-10-18 14:52:34 -07:00
Nick Terrell f45b157d95 Backport fix from commit 9e8b09a 
Fixes uninitialized memory reads.
Full commit hash: 9e8b09a7bd
 2016-10-18 14:22:49 -07:00
Yann Collet f7906d5955 detect early impossible decompression scenario in legacy decoder v0.5 2016-10-18 13:48:32 -07:00
Yann Collet 9313c8d953 no repeat mode in legacy v0.5 2016-10-18 13:36:15 -07:00
Yann Collet 83d7bdee4b fixed invalid invocation of dictionary in legacy decoder v0.5 2016-10-18 12:25:43 -07:00
Yann Collet 197a55ee7b fix edge case 2016-10-18 11:27:52 -07:00
Nick Terrell fd98087047 Fix stack buffer overflow in HUF_readCTable() 
If `w ==0` on line 153, then `CTable[n].nbBits == tableLog + 1`.
Then `nbPerRank[CTable[n].nbBits]` and `valPerRank[CTable[n].nbBits]`
are stack buffer overflows.
 2016-10-17 18:16:59 -07:00
Yann Collet 33fdd099bb fix command line interpretation 2016-10-17 17:48:48 -07:00
Yann Collet 06573e17be fixed minor corner case 2016-10-17 17:28:28 -07:00
Nick Terrell bfd943ace5 Fix buffer overrun in ZSTD_loadDictEntropyStats() 
The table log set by `FSE_readNCount()` was not checked in
`ZSTD_loadDictEntropyStats()`.  This caused `FSE_buildCTable()`
to stack/heap overflow in a few places.

The benchmarks look good, there is no obvious compression performance regression:

  > ./zstds/zstd.opt.0 -i10 -b1 -e10 ~/bench/silesia.tar
   1#silesia.tar       : 211988480 ->  73656930 (2.878), 271.6 MB/s , 716.8 MB/s
   2#silesia.tar       : 211988480 ->  70162842 (3.021), 204.8 MB/s , 671.1 MB/s
   3#silesia.tar       : 211988480 ->  66997986 (3.164), 156.8 MB/s , 658.6 MB/s
   4#silesia.tar       : 211988480 ->  66002591 (3.212), 136.4 MB/s , 665.3 MB/s
   5#silesia.tar       : 211988480 ->  65008480 (3.261),  98.9 MB/s , 647.0 MB/s
   6#silesia.tar       : 211988480 ->  62979643 (3.366),  65.2 MB/s , 670.4 MB/s
   7#silesia.tar       : 211988480 ->  61974560 (3.421),  44.9 MB/s , 688.2 MB/s
   8#silesia.tar       : 211988480 ->  61028308 (3.474),  32.4 MB/s , 711.9 MB/s
   9#silesia.tar       : 211988480 ->  60416751 (3.509),  21.1 MB/s , 718.1 MB/s
  10#silesia.tar       : 211988480 ->  60174239 (3.523),  22.2 MB/s , 721.8 MB/s

  > ./compress_zstds/zstd.opt.1 -i10 -b1 -e10 ~/bench/silesia.tar
   1#silesia.tar       : 211988480 ->  73656930 (2.878), 273.8 MB/s , 722.0 MB/s
   2#silesia.tar       : 211988480 ->  70162842 (3.021), 203.2 MB/s , 666.6 MB/s
   3#silesia.tar       : 211988480 ->  66997986 (3.164), 157.4 MB/s , 666.5 MB/s
   4#silesia.tar       : 211988480 ->  66002591 (3.212), 132.1 MB/s , 661.9 MB/s
   5#silesia.tar       : 211988480 ->  65008480 (3.261),  96.8 MB/s , 641.6 MB/s
   6#silesia.tar       : 211988480 ->  62979643 (3.366),  63.1 MB/s , 677.0 MB/s
   7#silesia.tar       : 211988480 ->  61974560 (3.421),  44.3 MB/s , 678.2 MB/s
   8#silesia.tar       : 211988480 ->  61028308 (3.474),  33.1 MB/s , 708.9 MB/s
   9#silesia.tar       : 211988480 ->  60416751 (3.509),  21.5 MB/s , 710.1 MB/s
  10#silesia.tar       : 211988480 ->  60174239 (3.523),  21.9 MB/s , 723.9 MB/s
 2016-10-17 16:55:52 -07:00
Nick Terrell 4db751668f Fix buffer overrun in ZSTD_loadEntropy() 
The table log set by `FSE_readNCount()` was not checked in
`ZSTD_loadEntropy()`.  This caused `FSE_buildDTable(dctx->MLTable, ...)`
to overwrite the beginning of `dctx->hufTable`.

The benchmarks look good, there is no obvious performance regression:

  > ./zstds/zstd.opt.0 -i10 -b1 -e5 ~/bench/silesia.tar
   1#silesia.tar       : 211988480 ->  73656930 (2.878), 268.2 MB/s , 701.0 MB/s
   2#silesia.tar       : 211988480 ->  70162842 (3.021), 199.5 MB/s , 666.9 MB/s
   3#silesia.tar       : 211988480 ->  66997986 (3.164), 154.9 MB/s , 655.6 MB/s
   4#silesia.tar       : 211988480 ->  66002591 (3.212), 128.9 MB/s , 648.4 MB/s
   5#silesia.tar       : 211988480 ->  65008480 (3.261),  98.4 MB/s , 633.4 MB/s

  > ./zstds/zstd.opt.2 -i10 -b1 -e5 ~/bench/silesia.tar
   1#silesia.tar       : 211988480 ->  73656930 (2.878), 266.1 MB/s , 703.7 MB/s
   2#silesia.tar       : 211988480 ->  70162842 (3.021), 199.0 MB/s , 666.6 MB/s
   3#silesia.tar       : 211988480 ->  66997986 (3.164), 156.2 MB/s , 656.2 MB/s
   4#silesia.tar       : 211988480 ->  66002591 (3.212), 133.2 MB/s , 647.4 MB/s
   5#silesia.tar       : 211988480 ->  65008480 (3.261),  96.3 MB/s , 633.3 MB/s
 2016-10-17 15:51:15 -07:00
Nick Terrell ccfcc643da Check if dict is empty before reading first byte 2016-10-17 11:46:03 -07:00