unclear or unsuitable licensing.
In general, we replace the files we can with liberally licensed data,
and remove all the others (in particular all the parts of the Canterbury
corpus that are not clearly in the public domain). The replacements
do not always have the exact same characteristics as the original ones,
but they are more than good enough to be useful for benchmarking.
git-svn-id: https://snappy.googlecode.com/svn/trunk@83 03e5f5b5-db94-4691-08a0-1a8bf15f6143
This is specifically motivated by DICOM's demands that embedded data
must be of an even number of bytes, but could in principle be used for
any sort of padding/alignment needed.
R=sanjay
git-svn-id: https://snappy.googlecode.com/svn/trunk@82 03e5f5b5-db94-4691-08a0-1a8bf15f6143
it would solve the problem if MSVC typically used autoconf. However, it gives
a natural place (config.h) to put the typedef even for MSVC.
R=jsbell
git-svn-id: https://snappy.googlecode.com/svn/trunk@80 03e5f5b5-db94-4691-08a0-1a8bf15f6143
backreference, make the signedness of the bytes produced clear,
by sticking it into a size_t. This avoids a signed/unsigned compare
warning from MSVC (public issue 71), and also is slightly clearer.
Since the line is now so long the explanatory comment about the -1u
trick has to go somewhere else anyway, I used the opportunity to
explain it in slightly more detail.
This is a purely stylistic change; the emitted assembler from GCC
is identical.
R=jeff
git-svn-id: https://snappy.googlecode.com/svn/trunk@79 03e5f5b5-db94-4691-08a0-1a8bf15f6143
whether there's 16 bytes free and then checking right afterwards
(when having subtracted the literal size) that there are now
5 bytes free, just check once for 21 bytes. This skips a compare
and a branch; although it is easily predictable, it is still
a few cycles on a fast path that we would like to get rid of.
Benchmarking this yields very confusing results. On open-source
GCC 4.8.1 on Haswell, we get exactly the expected results; the
benchmarks where we hit the fast path for literals (in particular
the two HTML benchmarks and the protobuf benchmark) give very nice
speedups, and the others are not really affected.
However, benchmarks with Google's GCC branch on other hardware
is much less clear. It seems that we have a weak loss in some cases
(and the win for the “typical” win cases are not nearly as clear),
but that it depends on microarchitecture and plain luck in how we run
the benchmark. Looking at the generated assembler, it seems that
the removal of the if causes other large-scale changes in how the
function is laid out, which makes it likely that this is just bad luck.
Thus, we should keep this change, even though its exact current impact is
unclear; it's a sensible change per se, and dropping it on the basis of
microoptimization for a given compiler (or even branch of a compiler)
would seem like a bad strategy in the long run.
Microbenchmark results (all in 64-bit, opt mode):
Nehalem, Google GCC:
Benchmark Base (ns) New (ns) Improvement
------------------------------------------------------------------------------
BM_UFlat/0 76747 75591 1.3GB/s html +1.5%
BM_UFlat/1 765756 757040 886.3MB/s urls +1.2%
BM_UFlat/2 10867 10893 10.9GB/s jpg -0.2%
BM_UFlat/3 124 131 1.4GB/s jpg_200 -5.3%
BM_UFlat/4 31663 31596 2.8GB/s pdf +0.2%
BM_UFlat/5 314162 308176 1.2GB/s html4 +1.9%
BM_UFlat/6 29668 29746 790.6MB/s cp -0.3%
BM_UFlat/7 12958 13386 796.4MB/s c -3.2%
BM_UFlat/8 3596 3682 966.0MB/s lsp -2.3%
BM_UFlat/9 1019193 1033493 953.3MB/s xls -1.4%
BM_UFlat/10 239 247 775.3MB/s xls_200 -3.2%
BM_UFlat/11 236411 240271 606.9MB/s txt1 -1.6%
BM_UFlat/12 206639 209768 571.2MB/s txt2 -1.5%
BM_UFlat/13 627803 635722 641.4MB/s txt3 -1.2%
BM_UFlat/14 845932 857816 538.2MB/s txt4 -1.4%
BM_UFlat/15 402107 391670 1.2GB/s bin +2.7%
BM_UFlat/16 283 279 683.6MB/s bin_200 +1.4%
BM_UFlat/17 46070 46815 781.5MB/s sum -1.6%
BM_UFlat/18 5053 5163 782.0MB/s man -2.1%
BM_UFlat/19 79721 76581 1.4GB/s pb +4.1%
BM_UFlat/20 251158 252330 697.5MB/s gaviota -0.5%
Sum of all benchmarks 4966150 4980396 -0.3%
Sandy Bridge, Google GCC:
Benchmark Base (ns) New (ns) Improvement
------------------------------------------------------------------------------
BM_UFlat/0 42850 42182 2.3GB/s html +1.6%
BM_UFlat/1 525660 515816 1.3GB/s urls +1.9%
BM_UFlat/2 7173 7283 16.3GB/s jpg -1.5%
BM_UFlat/3 92 91 2.1GB/s jpg_200 +1.1%
BM_UFlat/4 15147 14872 5.9GB/s pdf +1.8%
BM_UFlat/5 199936 192116 2.0GB/s html4 +4.1%
BM_UFlat/6 12796 12443 1.8GB/s cp +2.8%
BM_UFlat/7 6588 6400 1.6GB/s c +2.9%
BM_UFlat/8 2010 1951 1.8GB/s lsp +3.0%
BM_UFlat/9 761124 763049 1.3GB/s xls -0.3%
BM_UFlat/10 186 189 1016.1MB/s xls_200 -1.6%
BM_UFlat/11 159354 158460 918.6MB/s txt1 +0.6%
BM_UFlat/12 139732 139950 856.1MB/s txt2 -0.2%
BM_UFlat/13 429917 425027 961.7MB/s txt3 +1.2%
BM_UFlat/14 585255 587324 785.8MB/s txt4 -0.4%
BM_UFlat/15 276186 266173 1.8GB/s bin +3.8%
BM_UFlat/16 205 207 925.5MB/s bin_200 -1.0%
BM_UFlat/17 24925 24935 1.4GB/s sum -0.0%
BM_UFlat/18 2632 2576 1.5GB/s man +2.2%
BM_UFlat/19 40546 39108 2.8GB/s pb +3.7%
BM_UFlat/20 175803 168209 1048.9MB/s gaviota +4.5%
Sum of all benchmarks 3408117 3368361 +1.2%
Haswell, upstream GCC 4.8.1:
Benchmark Base (ns) New (ns) Improvement
------------------------------------------------------------------------------
BM_UFlat/0 46308 40641 2.3GB/s html +13.9%
BM_UFlat/1 513385 514706 1.3GB/s urls -0.3%
BM_UFlat/2 6197 6151 19.2GB/s jpg +0.7%
BM_UFlat/3 61 61 3.0GB/s jpg_200 +0.0%
BM_UFlat/4 13551 13429 6.5GB/s pdf +0.9%
BM_UFlat/5 198317 190243 2.0GB/s html4 +4.2%
BM_UFlat/6 14768 12560 1.8GB/s cp +17.6%
BM_UFlat/7 6453 6447 1.6GB/s c +0.1%
BM_UFlat/8 1991 1980 1.8GB/s lsp +0.6%
BM_UFlat/9 766947 770424 1.2GB/s xls -0.5%
BM_UFlat/10 170 169 1.1GB/s xls_200 +0.6%
BM_UFlat/11 164350 163554 888.7MB/s txt1 +0.5%
BM_UFlat/12 145444 143830 832.1MB/s txt2 +1.1%
BM_UFlat/13 437849 438413 929.2MB/s txt3 -0.1%
BM_UFlat/14 603587 605309 759.8MB/s txt4 -0.3%
BM_UFlat/15 249799 248067 1.9GB/s bin +0.7%
BM_UFlat/16 191 188 1011.4MB/s bin_200 +1.6%
BM_UFlat/17 26064 24778 1.4GB/s sum +5.2%
BM_UFlat/18 2620 2601 1.5GB/s man +0.7%
BM_UFlat/19 44551 37373 3.0GB/s pb +19.2%
BM_UFlat/20 165408 164584 1.0GB/s gaviota +0.5%
Sum of all benchmarks 3408011 3385508 +0.7%
git-svn-id: https://snappy.googlecode.com/svn/trunk@78 03e5f5b5-db94-4691-08a0-1a8bf15f6143
Windows does not have struct iovec defined anywhere,
so we define our own version that's equal to what UNIX
typically has.
The bulk of this patch was contributed by Mohit Aron.
R=jeff
git-svn-id: https://snappy.googlecode.com/svn/trunk@76 03e5f5b5-db94-4691-08a0-1a8bf15f6143
Previously, snappy_unittests would read from an absolute path /testdata/..;
convert it to use a relative path instead.
Patch from Marc-Antonie Ruel.
R=maruel
git-svn-id: https://snappy.googlecode.com/svn/trunk@72 03e5f5b5-db94-4691-08a0-1a8bf15f6143
use the LEA instruction more efficiently, since e.g. a + (b << 2) can be encoded
as one instruction. Even more importantly, it can constant-fold the
COPY_* enums together with the shifted negative constants, which also saves
some instructions. (We don't need it for LITERAL, since it happens to be 0.)
I am unsure why the compiler couldn't do this itself, but the theory is that
it cannot prove that len-1 and len-4 cannot underflow/wrap, and thus can't
do the optimization safely.
The gains are small but measurable; 0.5-1.0% over the BM_Z* benchmarks
(measured on Westmere, Sandy Bridge and Istanbul).
R=sanjay
git-svn-id: https://snappy.googlecode.com/svn/trunk@69 03e5f5b5-db94-4691-08a0-1a8bf15f6143
versions of libtool and automake on non-GNU platforms (e.g. Mac OS X).
R=sanjay
git-svn-id: https://snappy.googlecode.com/svn/trunk@68 03e5f5b5-db94-4691-08a0-1a8bf15f6143
it leaves the source in a state that's not appropriate for RawUncompress.
R=sanjay
git-svn-id: https://snappy.googlecode.com/svn/trunk@67 03e5f5b5-db94-4691-08a0-1a8bf15f6143
twice; it could cause division by zero in the unit test framework.
(We already had one fix for this in place, but it was incomplete.)
This could in theory happen on any system, since there are few guarantees
about gettimeofday(), but seems to only happen in practice on GNU/Hurd, where
gettimeofday() is cached and only updated ever so often.
R=sanjay
git-svn-id: https://snappy.googlecode.com/svn/trunk@65 03e5f5b5-db94-4691-08a0-1a8bf15f6143
apparently Debian still targets these by default, giving us segfaults on
armel.
R=sanjay
git-svn-id: https://snappy.googlecode.com/svn/trunk@64 03e5f5b5-db94-4691-08a0-1a8bf15f6143
Achieved by moving logging macro definitions to a test-only
header file, and by changing non-test code to use assert,
fprintf, and abort instead of LOG/CHECK macros.
R=sesse
git-svn-id: https://snappy.googlecode.com/svn/trunk@62 03e5f5b5-db94-4691-08a0-1a8bf15f6143
to 2 MB. This fixes issues with running the unit test on platforms with
little RAM (e.g. some ARM boards).
Also, reactivate the 2 MB test for 64-bit platforms; there's no good
reason why it shouldn't be.
R=sanjay
git-svn-id: https://snappy.googlecode.com/svn/trunk@58 03e5f5b5-db94-4691-08a0-1a8bf15f6143
warnings. There are still some in the unit test, but the main .cc file should
be clean. We haven't enabled -Wall for the default build, since the unit test
is still not clean.
This also fixes a real bug in the open-source implementation of
ReadFileToStringOrDie(); it would not detect errors correctly.
I had to go through some pains to avoid performance loss as the types
were changed; I think there might still be some with 32-bit if and only if LFS
is enabled (ie., size_t is 64-bit), but for regular 32-bit and 64-bit I can't
see any losses, and I've diffed the generated GCC assembler between the old and
new code without seeing any significant choices. If anything, it's ever so
slightly faster.
This may or may not enable compression of very large blocks (>2^32 bytes)
when size_t is 64-bit, but I haven't checked, and it is still not a supported
case.
git-svn-id: https://snappy.googlecode.com/svn/trunk@56 03e5f5b5-db94-4691-08a0-1a8bf15f6143
the current point, but there seems to be enough of a general interest in the
topic (cf. public bug #34).
R=csilvers,sanjay
git-svn-id: https://snappy.googlecode.com/svn/trunk@55 03e5f5b5-db94-4691-08a0-1a8bf15f6143
enum declarations, as it seems the Sun compiler does not like it.
Based on patch by Travis Vitek.
git-svn-id: https://snappy.googlecode.com/svn/trunk@48 03e5f5b5-db94-4691-08a0-1a8bf15f6143
instead of "4-byte offset" for the longest copies.
Also fix an inconsistency in the heading for section 2.2.3.
Both patches by Patrick Pelletier.
R=csilvers
git-svn-id: https://snappy.googlecode.com/svn/trunk@45 03e5f5b5-db94-4691-08a0-1a8bf15f6143
identical, even regarding cv-qualifiers.
This is required to work around a bug in the Solaris Studio C++ compiler
(it does not properly disregard cv-qualifiers when doing name mangling).
R=sanjay
git-svn-id: https://snappy.googlecode.com/svn/trunk@44 03e5f5b5-db94-4691-08a0-1a8bf15f6143
(I stumbled into this when changing the way we decompress literals.)
R=csilvers
Revision created by MOE tool push_codebase.
git-svn-id: https://snappy.googlecode.com/svn/trunk@43 03e5f5b5-db94-4691-08a0-1a8bf15f6143
Whenever we try to enter a copy fast-path, there is a certain cost in checking
that all the preconditions are in place, but it's normally offset by the fact
that we can usually take the cheaper path. However, in a certain path we've
already established that "avail < literal_length", which usually means that
either the available space is small, or the literal is big. Both will disqualify
us from taking the fast path, and thus we take the hit from the precondition
checking without gaining much from having a fast path. Thus, simply don't try
the fast path in this situation -- we're already on a slow path anyway
(one where we need to refill more data from the reader).
I'm a bit surprised at how much this gained; it could be that this path is
more common than I thought, or that the simpler structure somehow makes the
compiler happier. I haven't looked at the assembler, but it's a win across
the board on both Core 2, Core i7 and Opteron, at least for the cases we
typically care about. The gains seem to be the largest on Core i7, though.
Results from my Core i7 workstation:
Benchmark Time(ns) CPU(ns) Iterations
---------------------------------------------------
BM_UFlat/0 73337 73091 190996 1.3GB/s html [ +1.7%]
BM_UFlat/1 696379 693501 20173 965.5MB/s urls [ +2.7%]
BM_UFlat/2 9765 9734 1472135 12.1GB/s jpg [ +0.7%]
BM_UFlat/3 29720 29621 472973 3.0GB/s pdf [ +1.8%]
BM_UFlat/4 294636 293834 47782 1.3GB/s html4 [ +2.3%]
BM_UFlat/5 28399 28320 494700 828.5MB/s cp [ +3.5%]
BM_UFlat/6 12795 12760 1000000 833.3MB/s c [ +1.2%]
BM_UFlat/7 3984 3973 3526448 893.2MB/s lsp [ +5.7%]
BM_UFlat/8 991996 989322 14141 992.6MB/s xls [ +3.3%]
BM_UFlat/9 228620 227835 61404 636.6MB/s txt1 [ +4.0%]
BM_UFlat/10 197114 196494 72165 607.5MB/s txt2 [ +3.5%]
BM_UFlat/11 605240 603437 23217 674.4MB/s txt3 [ +3.7%]
BM_UFlat/12 804157 802016 17456 573.0MB/s txt4 [ +3.9%]
BM_UFlat/13 347860 346998 40346 1.4GB/s bin [ +1.2%]
BM_UFlat/14 44684 44559 315315 818.4MB/s sum [ +2.3%]
BM_UFlat/15 5120 5106 2739726 789.4MB/s man [ +3.3%]
BM_UFlat/16 76591 76355 183486 1.4GB/s pb [ +2.8%]
BM_UFlat/17 238564 237828 58824 739.1MB/s gaviota [ +1.6%]
BM_UValidate/0 42194 42060 333333 2.3GB/s html [ -0.1%]
BM_UValidate/1 433182 432005 32407 1.5GB/s urls [ -0.1%]
BM_UValidate/2 197 196 71428571 603.3GB/s jpg [ +0.5%]
BM_UValidate/3 14494 14462 972222 6.1GB/s pdf [ +0.5%]
BM_UValidate/4 168444 167836 83832 2.3GB/s html4 [ +0.1%]
R=jeff
Revision created by MOE tool push_codebase.
git-svn-id: https://snappy.googlecode.com/svn/trunk@42 03e5f5b5-db94-4691-08a0-1a8bf15f6143
Looking up into and decoding the values from char_table has long shown up as a
hotspot in the decompressor. While it turns out that it's hard to make a more
efficient decoder for the copy ops, the literals are simple enough that we can
decode them without needing a table lookup. (This means that 1/4 of the table
is now unused, although that in itself doesn't buy us anything.)
The gains are small, but definitely present; some tests win as much as 10%,
but 1-4% is more typical. These results are from Core i7, in 64-bit mode;
Core 2 and Opteron show similar results. (I've run with more iterations
than unusual to make sure the smaller gains don't drown entirely in noise.)
Benchmark Time(ns) CPU(ns) Iterations
---------------------------------------------------
BM_UFlat/0 74665 74428 182055 1.3GB/s html [ +3.1%]
BM_UFlat/1 714106 711997 19663 940.4MB/s urls [ +4.4%]
BM_UFlat/2 9820 9789 1427115 12.1GB/s jpg [ -1.2%]
BM_UFlat/3 30461 30380 465116 2.9GB/s pdf [ +0.8%]
BM_UFlat/4 301445 300568 46512 1.3GB/s html4 [ +2.2%]
BM_UFlat/5 29338 29263 479452 801.8MB/s cp [ +1.6%]
BM_UFlat/6 13004 12970 1000000 819.9MB/s c [ +2.1%]
BM_UFlat/7 4180 4168 3349282 851.4MB/s lsp [ +1.3%]
BM_UFlat/8 1026149 1024000 10000 959.0MB/s xls [+10.7%]
BM_UFlat/9 237441 236830 59072 612.4MB/s txt1 [ +0.3%]
BM_UFlat/10 203966 203298 69307 587.2MB/s txt2 [ +0.8%]
BM_UFlat/11 627230 625000 22400 651.2MB/s txt3 [ +0.7%]
BM_UFlat/12 836188 833979 16787 551.0MB/s txt4 [ +1.3%]
BM_UFlat/13 351904 350750 39886 1.4GB/s bin [ +3.8%]
BM_UFlat/14 45685 45562 308370 800.4MB/s sum [ +5.9%]
BM_UFlat/15 5286 5270 2656546 764.9MB/s man [ +1.5%]
BM_UFlat/16 78774 78544 178117 1.4GB/s pb [ +4.3%]
BM_UFlat/17 242270 241345 58091 728.3MB/s gaviota [ +1.2%]
BM_UValidate/0 42149 42000 333333 2.3GB/s html [ -3.0%]
BM_UValidate/1 432741 431303 32483 1.5GB/s urls [ +7.8%]
BM_UValidate/2 198 197 71428571 600.7GB/s jpg [+16.8%]
BM_UValidate/3 14560 14521 965517 6.1GB/s pdf [ -4.1%]
BM_UValidate/4 169065 168671 83832 2.3GB/s html4 [ -2.9%]
R=jeff
Revision created by MOE tool push_codebase.
git-svn-id: https://snappy.googlecode.com/svn/trunk@41 03e5f5b5-db94-4691-08a0-1a8bf15f6143
state of ip_ after decompression (or attempted decompresion) is
completely irrelevant, so we don't need the trailer.
Performance is, as expected, mostly flat -- there's a curious ~3-5%
loss in the "lsp" test, but that test case is so short it is hard to say
anything definitive about why (most likely, it's some sort of
unrelated effect).
R=jeff
git-svn-id: https://snappy.googlecode.com/svn/trunk@39 03e5f5b5-db94-4691-08a0-1a8bf15f6143
It is seemingly hard for the compiler to understand that ip_, the current input
pointer into the compressed data stream, can not alias on anything else, and
thus using it directly will incur memory traffic as it cannot be kept in a
register. The code already knew about this and cached it into a local
variable, but since Step() only decoded one tag, it had to move ip_ back into
place between every tag. This seems to have cost us a significant amount of
performance, so changing Step() into a function that decodes as much as it can
before it saves ip_ back and returns. (Note that Step() was already inlined,
so it is not the manual inlining that buys the performance here.)
The wins are about 3-6% for Core 2, 6-13% on Core i7 and 5-12% on Opteron
(for plain array-to-array decompression, in 64-bit opt mode).
There is a tiny difference in the behavior here; if an invalid literal is
encountered (ie., the writer refuses the Append() operation), ip_ will now
point to the byte past the tag byte, instead of where the literal was
originally thought to end. However, we don't use ip_ for anything after
DecompressAllTags() has returned, so this should not change external behavior
in any way.
Microbenchmark results for Core i7, 64-bit (Opteron results are similar):
Benchmark Time(ns) CPU(ns) Iterations
---------------------------------------------------
BM_UFlat/0 79134 79110 8835 1.2GB/s html [ +6.2%]
BM_UFlat/1 786126 786096 891 851.8MB/s urls [+10.0%]
BM_UFlat/2 9948 9948 69125 11.9GB/s jpg [ -1.3%]
BM_UFlat/3 31999 31998 21898 2.7GB/s pdf [ +6.5%]
BM_UFlat/4 318909 318829 2204 1.2GB/s html4 [ +6.5%]
BM_UFlat/5 31384 31390 22363 747.5MB/s cp [ +9.2%]
BM_UFlat/6 14037 14034 49858 757.7MB/s c [+10.6%]
BM_UFlat/7 4612 4612 151395 769.5MB/s lsp [ +9.5%]
BM_UFlat/8 1203174 1203007 582 816.3MB/s xls [+19.3%]
BM_UFlat/9 253869 253955 2757 571.1MB/s txt1 [+11.4%]
BM_UFlat/10 219292 219290 3194 544.4MB/s txt2 [+12.1%]
BM_UFlat/11 672135 672131 1000 605.5MB/s txt3 [+11.2%]
BM_UFlat/12 902512 902492 776 509.2MB/s txt4 [+12.5%]
BM_UFlat/13 372110 371998 1881 1.3GB/s bin [ +5.8%]
BM_UFlat/14 50407 50407 10000 723.5MB/s sum [+13.5%]
BM_UFlat/15 5699 5701 100000 707.2MB/s man [+12.4%]
BM_UFlat/16 83448 83424 8383 1.3GB/s pb [ +5.7%]
BM_UFlat/17 256958 256963 2723 684.1MB/s gaviota [ +7.9%]
BM_UValidate/0 42795 42796 16351 2.2GB/s html [+25.8%]
BM_UValidate/1 490672 490622 1427 1.3GB/s urls [+22.7%]
BM_UValidate/2 237 237 2950297 499.0GB/s jpg [+24.9%]
BM_UValidate/3 14610 14611 47901 6.0GB/s pdf [+26.8%]
BM_UValidate/4 171973 171990 4071 2.2GB/s html4 [+25.7%]
git-svn-id: https://snappy.googlecode.com/svn/trunk@38 03e5f5b5-db94-4691-08a0-1a8bf15f6143
