diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2021-11-13 15:32:30 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2021-11-13 15:32:30 -0800 |
commit | c8c109546a19613d323a319d0c921cb1f317e629 (patch) | |
tree | 5cf721ff53b9caba1162065cd16071e5226d40db /lib/zstd/compress/zstd_double_fast.c | |
parent | ccfff0a2bd2a30de130b5623d242ddecd0272bc2 (diff) | |
parent | 0a8ea235837cc39f27c45689930aa97ae91d5953 (diff) |
Merge tag 'zstd-for-linus-v5.16' of git://github.com/terrelln/linux
Pull zstd update from Nick Terrell:
"Update to zstd-1.4.10.
Add myself as the maintainer of zstd and update the zstd version in
the kernel, which is now 4 years out of date, to a much more recent
zstd release. This includes bug fixes, much more extensive fuzzing,
and performance improvements. And generates the kernel zstd
automatically from upstream zstd, so it is easier to keep the zstd
verison up to date, and we don't fall so far out of date again.
This includes 5 commits that update the zstd library version:
- Adds a new kernel-style wrapper around zstd.
This wrapper API is functionally equivalent to the subset of the
current zstd API that is currently used. The wrapper API changes to
be kernel style so that the symbols don't collide with zstd's
symbols. The update to zstd-1.4.10 maintains the same API and
preserves the semantics, so that none of the callers need to be
updated. All callers are updated in the commit, because there are
zero functional changes.
- Adds an indirection for `lib/decompress_unzstd.c` so it doesn't
depend on the layout of `lib/zstd/` to include every source file.
This allows the next patch to be automatically generated.
- Imports the zstd-1.4.10 source code. This commit is automatically
generated from upstream zstd (https://github.com/facebook/zstd).
- Adds me (terrelln@fb.com) as the maintainer of `lib/zstd`.
- Fixes a newly added build warning for clang.
The discussion around this patchset has been pretty long, so I've
included a FAQ-style summary of the history of the patchset, and why
we are taking this approach.
Why do we need to update?
-------------------------
The zstd version in the kernel is based off of zstd-1.3.1, which is
was released August 20, 2017. Since then zstd has seen many bug fixes
and performance improvements. And, importantly, upstream zstd is
continuously fuzzed by OSS-Fuzz, and bug fixes aren't backported to
older versions. So the only way to sanely get these fixes is to keep
up to date with upstream zstd.
There are no known security issues that affect the kernel, but we need
to be able to update in case there are. And while there are no known
security issues, there are relevant bug fixes. For example the problem
with large kernel decompression has been fixed upstream for over 2
years [1]
Additionally the performance improvements for kernel use cases are
significant. Measured for x86_64 on my Intel i9-9900k @ 3.6 GHz:
- BtrFS zstd compression at levels 1 and 3 is 5% faster
- BtrFS zstd decompression+read is 15% faster
- SquashFS zstd decompression+read is 15% faster
- F2FS zstd compression+write at level 3 is 8% faster
- F2FS zstd decompression+read is 20% faster
- ZRAM decompression+read is 30% faster
- Kernel zstd decompression is 35% faster
- Initramfs zstd decompression+build is 5% faster
On top of this, there are significant performance improvements coming
down the line in the next zstd release, and the new automated update
patch generation will allow us to pull them easily.
How is the update patch generated?
----------------------------------
The first two patches are preparation for updating the zstd version.
Then the 3rd patch in the series imports upstream zstd into the
kernel. This patch is automatically generated from upstream. A script
makes the necessary changes and imports it into the kernel. The
changes are:
- Replace all libc dependencies with kernel replacements and rewrite
includes.
- Remove unncessary portability macros like: #if defined(_MSC_VER).
- Use the kernel xxhash instead of bundling it.
This automation gets tested every commit by upstream's continuous
integration. When we cut a new zstd release, we will submit a patch to
the kernel to update the zstd version in the kernel.
The automated process makes it easy to keep the kernel version of zstd
up to date. The current zstd in the kernel shares the guts of the
code, but has a lot of API and minor changes to work in the kernel.
This is because at the time upstream zstd was not ready to be used in
the kernel envrionment as-is. But, since then upstream zstd has
evolved to support being used in the kernel as-is.
Why are we updating in one big patch?
-------------------------------------
The 3rd patch in the series is very large. This is because it is
restructuring the code, so it both deletes the existing zstd, and
re-adds the new structure. Future updates will be directly
proportional to the changes in upstream zstd since the last import.
They will admittidly be large, as zstd is an actively developed
project, and has hundreds of commits between every release. However,
there is no other great alternative.
One option ruled out is to replay every upstream zstd commit. This is
not feasible for several reasons:
- There are over 3500 upstream commits since the zstd version in the
kernel.
- The automation to automatically generate the kernel update was only
added recently, so older commits cannot easily be imported.
- Not every upstream zstd commit builds.
- Only zstd releases are "supported", and individual commits may have
bugs that were fixed before a release.
Another option to reduce the patch size would be to first reorganize
to the new file structure, and then apply the patch. However, the
current kernel zstd is formatted with clang-format to be more
"kernel-like". But, the new method imports zstd as-is, without
additional formatting, to allow for closer correlation with upstream,
and easier debugging. So the patch wouldn't be any smaller.
It also doesn't make sense to import upstream zstd commit by commit
going forward. Upstream zstd doesn't support production use cases
running of the development branch. We have a lot of post-commit
fuzzing that catches many bugs, so indiviudal commits may be buggy,
but fixed before a release. So going forward, I intend to import every
(important) zstd release into the Kernel.
So, while it isn't ideal, updating in one big patch is the only patch
I see forward.
Who is responsible for this code?
---------------------------------
I am. This patchset adds me as the maintainer for zstd. Previously,
there was no tree for zstd patches. Because of that, there were
several patches that either got ignored, or took a long time to merge,
since it wasn't clear which tree should pick them up. I'm officially
stepping up as maintainer, and setting up my tree as the path through
which zstd patches get merged. I'll make sure that patches to the
kernel zstd get ported upstream, so they aren't erased when the next
version update happens.
How is this code tested?
------------------------
I tested every caller of zstd on x86_64 (BtrFS, ZRAM, SquashFS, F2FS,
Kernel, InitRAMFS). I also tested Kernel & InitRAMFS on i386 and
aarch64. I checked both performance and correctness.
Also, thanks to many people in the community who have tested these
patches locally.
Lastly, this code will bake in linux-next before being merged into
v5.16.
Why update to zstd-1.4.10 when zstd-1.5.0 has been released?
------------------------------------------------------------
This patchset has been outstanding since 2020, and zstd-1.4.10 was the
latest release when it was created. Since the update patch is
automatically generated from upstream, I could generate it from
zstd-1.5.0.
However, there were some large stack usage regressions in zstd-1.5.0,
and are only fixed in the latest development branch. And the latest
development branch contains some new code that needs to bake in the
fuzzer before I would feel comfortable releasing to the kernel.
Once this patchset has been merged, and we've released zstd-1.5.1, we
can update the kernel to zstd-1.5.1, and exercise the update process.
You may notice that zstd-1.4.10 doesn't exist upstream. This release
is an artifical release based off of zstd-1.4.9, with some fixes for
the kernel backported from the development branch. I will tag the
zstd-1.4.10 release after this patchset is merged, so the Linux Kernel
is running a known version of zstd that can be debugged upstream.
Why was a wrapper API added?
----------------------------
The first versions of this patchset migrated the kernel to the
upstream zstd API. It first added a shim API that supported the new
upstream API with the old code, then updated callers to use the new
shim API, then transitioned to the new code and deleted the shim API.
However, Cristoph Hellwig suggested that we transition to a kernel
style API, and hide zstd's upstream API behind that. This is because
zstd's upstream API is supports many other use cases, and does not
follow the kernel style guide, while the kernel API is focused on the
kernel's use cases, and follows the kernel style guide.
Where is the previous discussion?
---------------------------------
Links for the discussions of the previous versions of the patch set
below. The largest changes in the design of the patchset are driven by
the discussions in v11, v5, and v1. Sorry for the mix of links, I
couldn't find most of the the threads on lkml.org"
Link: https://lkml.org/lkml/2020/9/29/27 [1]
Link: https://www.spinics.net/lists/linux-crypto/msg58189.html [v12]
Link: https://lore.kernel.org/linux-btrfs/20210430013157.747152-1-nickrterrell@gmail.com/ [v11]
Link: https://lore.kernel.org/lkml/20210426234621.870684-2-nickrterrell@gmail.com/ [v10]
Link: https://lore.kernel.org/linux-btrfs/20210330225112.496213-1-nickrterrell@gmail.com/ [v9]
Link: https://lore.kernel.org/linux-f2fs-devel/20210326191859.1542272-1-nickrterrell@gmail.com/ [v8]
Link: https://lkml.org/lkml/2020/12/3/1195 [v7]
Link: https://lkml.org/lkml/2020/12/2/1245 [v6]
Link: https://lore.kernel.org/linux-btrfs/20200916034307.2092020-1-nickrterrell@gmail.com/ [v5]
Link: https://www.spinics.net/lists/linux-btrfs/msg105783.html [v4]
Link: https://lkml.org/lkml/2020/9/23/1074 [v3]
Link: https://www.spinics.net/lists/linux-btrfs/msg105505.html [v2]
Link: https://lore.kernel.org/linux-btrfs/20200916034307.2092020-1-nickrterrell@gmail.com/ [v1]
Signed-off-by: Nick Terrell <terrelln@fb.com>
Tested By: Paul Jones <paul@pauljones.id.au>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com> # LLVM/Clang v13.0.0 on x86-64
Tested-by: Jean-Denis Girard <jd.girard@sysnux.pf>
* tag 'zstd-for-linus-v5.16' of git://github.com/terrelln/linux:
lib: zstd: Add cast to silence clang's -Wbitwise-instead-of-logical
MAINTAINERS: Add maintainer entry for zstd
lib: zstd: Upgrade to latest upstream zstd version 1.4.10
lib: zstd: Add decompress_sources.h for decompress_unzstd
lib: zstd: Add kernel-specific API
Diffstat (limited to 'lib/zstd/compress/zstd_double_fast.c')
-rw-r--r-- | lib/zstd/compress/zstd_double_fast.c | 519 |
1 files changed, 519 insertions, 0 deletions
diff --git a/lib/zstd/compress/zstd_double_fast.c b/lib/zstd/compress/zstd_double_fast.c new file mode 100644 index 000000000000..b0424d23ac57 --- /dev/null +++ b/lib/zstd/compress/zstd_double_fast.c @@ -0,0 +1,519 @@ +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * 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" + + +void ZSTD_fillDoubleHashTable(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; + } } +} + + +FORCE_INLINE_TEMPLATE +size_t ZSTD_compressBlock_doubleFast_generic( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize, + U32 const mls /* template */, ZSTD_dictMode_e const dictMode) +{ + 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]; + U32 offsetSaved = 0; + + const ZSTD_matchState_t* const dms = ms->dictMatchState; + const ZSTD_compressionParameters* const dictCParams = + dictMode == ZSTD_dictMatchState ? + &dms->cParams : NULL; + const U32* const dictHashLong = dictMode == ZSTD_dictMatchState ? + dms->hashTable : NULL; + const U32* const dictHashSmall = dictMode == ZSTD_dictMatchState ? + dms->chainTable : NULL; + const U32 dictStartIndex = dictMode == ZSTD_dictMatchState ? + dms->window.dictLimit : 0; + const BYTE* const dictBase = dictMode == ZSTD_dictMatchState ? + dms->window.base : NULL; + const BYTE* const dictStart = dictMode == ZSTD_dictMatchState ? + dictBase + dictStartIndex : NULL; + const BYTE* const dictEnd = dictMode == ZSTD_dictMatchState ? + dms->window.nextSrc : NULL; + const U32 dictIndexDelta = dictMode == ZSTD_dictMatchState ? + prefixLowestIndex - (U32)(dictEnd - dictBase) : + 0; + const U32 dictHBitsL = dictMode == ZSTD_dictMatchState ? + dictCParams->hashLog : hBitsL; + const U32 dictHBitsS = dictMode == ZSTD_dictMatchState ? + dictCParams->chainLog : hBitsS; + const U32 dictAndPrefixLength = (U32)((ip - prefixLowest) + (dictEnd - dictStart)); + + DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_generic"); + + assert(dictMode == ZSTD_noDict || dictMode == ZSTD_dictMatchState); + + /* if a dictionary is attached, it must be within window range */ + if (dictMode == ZSTD_dictMatchState) { + assert(ms->window.dictLimit + (1U << cParams->windowLog) >= endIndex); + } + + /* init */ + ip += (dictAndPrefixLength == 0); + if (dictMode == ZSTD_noDict) { + U32 const curr = (U32)(ip - base); + U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, cParams->windowLog); + U32 const maxRep = curr - windowLow; + if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0; + if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0; + } + if (dictMode == ZSTD_dictMatchState) { + /* 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 dictHL = ZSTD_hashPtr(ip, dictHBitsL, 8); + size_t const dictHS = ZSTD_hashPtr(ip, dictHBitsS, mls); + 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 = (dictMode == ZSTD_dictMatchState + && repIndex < prefixLowestIndex) ? + dictBase + (repIndex - dictIndexDelta) : + base + repIndex; + hashLong[h2] = hashSmall[h] = curr; /* update hash tables */ + + /* check dictMatchState repcode */ + if (dictMode == ZSTD_dictMatchState + && ((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, 0, mLength-MINMATCH); + goto _match_stored; + } + + /* check noDict repcode */ + if ( dictMode == ZSTD_noDict + && ((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, 0, mLength-MINMATCH); + 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 (dictMode == ZSTD_dictMatchState) { + /* check dictMatchState long match */ + U32 const dictMatchIndexL = dictHashLong[dictHL]; + 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 (dictMode == ZSTD_dictMatchState) { + /* check dictMatchState short match */ + U32 const dictMatchIndexS = dictHashSmall[dictHS]; + 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 dictHLNext = ZSTD_hashPtr(ip+1, dictHBitsL, 8); + U32 const matchIndexL3 = hashLong[hl3]; + 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 (dictMode == ZSTD_dictMatchState) { + /* check dict long +1 match */ + U32 const dictMatchIndexL3 = dictHashLong[dictHLNext]; + 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 (dictMode == ZSTD_dictMatchState && 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 + ZSTD_REP_MOVE, mLength-MINMATCH); + +_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 */ + if (dictMode == ZSTD_dictMatchState) { + while (ip <= ilimit) { + U32 const current2 = (U32)(ip-base); + U32 const repIndex2 = current2 - offset_2; + const BYTE* repMatch2 = dictMode == ZSTD_dictMatchState + && 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, 0, repLength2-MINMATCH); + hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2; + hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2; + ip += repLength2; + anchor = ip; + continue; + } + break; + } } + + if (dictMode == ZSTD_noDict) { + 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, 0, rLength-MINMATCH); + ip += rLength; + anchor = ip; + continue; /* faster when present ... (?) */ + } } } + } /* while (ip < ilimit) */ + + /* save reps for next block */ + rep[0] = offset_1 ? offset_1 : offsetSaved; + rep[1] = offset_2 ? offset_2 : offsetSaved; + + /* Return the last literals size */ + return (size_t)(iend - anchor); +} + + +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_generic(ms, seqStore, rep, src, srcSize, 4, ZSTD_noDict); + case 5 : + return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 5, ZSTD_noDict); + case 6 : + return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 6, ZSTD_noDict); + case 7 : + return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 7, ZSTD_noDict); + } +} + + +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_generic(ms, seqStore, rep, src, srcSize, 4, ZSTD_dictMatchState); + case 5 : + return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 5, ZSTD_dictMatchState); + case 6 : + return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 6, ZSTD_dictMatchState); + case 7 : + return ZSTD_compressBlock_doubleFast_generic(ms, seqStore, rep, src, srcSize, 7, ZSTD_dictMatchState); + } +} + + +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_generic(ms, seqStore, rep, src, srcSize, mls, ZSTD_noDict); + + /* 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 */ + & (repIndex > dictStartIndex)) + && (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, 0, mLength-MINMATCH); + } 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 + ZSTD_REP_MOVE, mLength-MINMATCH); + + } 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 + ZSTD_REP_MOVE, mLength-MINMATCH); + + } 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 */ + & (repIndex2 > 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, 0, repLength2-MINMATCH); + 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); +} + + +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_generic(ms, seqStore, rep, src, srcSize, 4); + case 5 : + return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 5); + case 6 : + return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 6); + case 7 : + return ZSTD_compressBlock_doubleFast_extDict_generic(ms, seqStore, rep, src, srcSize, 7); + } +} |