This commit improves the C and SSE2 Huffman encoding implementations in
the following ways:
- Avoid using xmm8-xmm15 in the x86-64 SSE2 implementation. There is no
actual need to use those registers, and avoiding them produces a
cleaner WIN64 function entry/exit-- as well as shorter code, since REX
prefixes can be avoided (this is helpful on certain CPUs, such as
Intel Atom, for which instruction fetch and decoding can be a
bottleneck.)
- Optimize register usage so that fewer REX prefixes and
register-register moves are needed.
- Use the bit counter to store the number of free bits in the bit buffer
rather than the number of bits in the bit buffer. This changes the
method for inserting a code into the bit buffer to:
(put_buffer |= code << (free_bits -= code_size));
As a result:
* Only one bit counter needs to stay in a register (we just keep it in
cl.)
* The bit buffer contents are already properly aligned to be written
out (after a byte swap.)
* Adjusting the free bits counter and checking if the bit buffer is
full can be combined into a single operation.
* We can wait to flush the bit buffer until the buffer is actually
full and not just in danger of becoming full. Thus, eight bytes can
be flushed at a time.
- Speed is quite sensitive to the alignment of branch target labels, so
insert some padding and remove branches from the flush code.
(Flushing this way isn't actually faster when compared to using
branches, but the branchless code doesn't need extra alignment and is
thus smaller.)
- Speculatively write out the bit buffer as a single 8-byte write,
falling back to a byte-by-byte write only if there are any 0xFF bytes
in the bit buffer that need to be encoded as 0xFF 0x00.
- Use MMX registers for the 32-bit implementation (so the bit buffer can
be 64 bits wide.)
- Slightly reduce overall function code size.
- Eliminate or combine a few SSE instructions.
- Make some minor improvements to instruction scheduling.
- Adjust flush_bits() in jchuff.c to handle cases in which the bit
buffer has less than 7 free bits (apparently that couldn't happen
before.)
Based on:
947a09defa262ebb6b816e9a091221
See change log for performance claims.
Closes#292
byte, word, dword, qword, oword, and yword are all assembler keywords,
so it makes sense to use lowercase for these so as not to mistake them
for macros or constants.
(regression introduced by 5b177b3cab)
The SSE2 implementation of progressive Huffman encoding performed
extraneous iterations when the scan length was a multiple of 16.
Based on:
bb7f1ef983Fixes#335Closes#367
According to Intel's manual [1], "If a value entered for CPUID.EAX is
higher than the maximum input value for basic or extended function for
that processor then the data for the highest basic information leaf is
returned."
Right now, libjpeg-turbo doesn't first check that leaf 07H is supported
before attempting to use it, so the ostensible AVX2 bit (Bit 05) of the
CPUID result might actually be Bit 05 from a lower leaf. That bit might
be set, even if the CPU doesn't support AVX2.
This commit modifies the x86 and x86-64 SIMD feature detection code so
that it first checks whether CPUID leaf 07H is supported before
attempting to use it to check for AVX2 instruction support.
DRC:
This commit should fix
https://bugzilla.mozilla.org/show_bug.cgi?id=1520760
However, I have not personally been able to reproduce that issue,
despite using a Nehalem (pre-AVX2) CPU on which the maximum CPUID leaf
has been limited via a BIOS setting.
Closes#348
[1]
"Intel® 64 and IA-32 Architectures Software Developer's Manual, Volume 2 (2A, 2B, 2C & 2D): Instruction Set Reference, A-Z", https://software.intel.com/sites/default/files/managed/a4/60/325383-sdm-vol-2abcd.pdf, page 3-192.
When simd/arm/jsimd.c is compiled with __ARM_NEON__ defined (which will
be the case if -mfpu=neon is passed to the compiler), the
parse_proc_cpuinfo() and check_feature() functions and the bufsize
variable are unused and thus need to be #ifdef'ed out in order to avoid
compiler warnings. Note that the bufsize variable was already #ifdef'ed
out on Linux but not on Android due to lack of parentheses (&& takes
precedence over ||.)
Closes#331
We have more than eight registers to work with, as well as three-operand
intrinsics, so there's no need for the implementation to be such a
literal port of the MMX code.
libjpeg-turbo has never really supported such compilers, since (AFAIK)
they are non-existent on any modern computing platform and thus
impossible for us to test. (Also, the TurboJPEG API would break without
unsigned chars.)
Furthermore, the unified CMake-based build system introduced in 2.0
always defines HAVE_UNSIGNED_CHAR, so retaining other code paths is
pointless. Eliminating support for compilers without unsigned char
eliminates the need for the GETJSAMPLE() macro, which improves the
readability of many parts of the code as well as improving the
performance of writing Targa and Windows BMP files.
Fixes#317
Apparently Windows 7 without SP1 has O/S support for XSAVE but not for
YMM registers, and this exposed a bug in our usage of xgetbv. The test
instruction will set ZF only if none of the bits match between the two
operarands, so in effect, we were enabling AVX2 instructions if the O/S
supported XSAVE and the CPU supported AVX2 but the O/S only supported
XMM registers. This bug was not exposed on, for instance, Windows XP or
RHEL 5 because those O/S's do not support XSAVE.
Fixes#288
The DSPr2 extensions have been verified to work with little endian MIPS.
Whether or not CMAKE_SYSTEM_PROCESSOR is set to "mips" or "mipsel" in a
little endian MIPS environment seems to be inconsistent, but our build
system needs to handle both cases.
(for instance, when passing -msoft-float to the compiler)
The instructions used by jsimd_quantize_float_dspr2() and
jsimd_convsamp_float_dspr2() don't work with the soft float ABI, so
disable those functions when soft float is enabled.
Based on:
129a739bfaCloses#272
This is basically the same test that was performed in acinclude.m4 in
the old autotools-based build system. It was not ported to the
CMake-based build system because I previously had no way of testing
a non-DSPr2 build environment.
Fixes#248
The old Un*x (autotools-based) build system always auto-generated this
file, but that behavior was more or less a relic of the days before the
libjpeg-turbo colorspace extensions were implemented. The thinking was
that, if a particular developer wanted to change RGB_RED, RGB_GREEN,
RGB_BLUE, or RGB_PIXELSIZE in order to compress from/decompress to
different RGB pixel layouts, then the SIMD extensions should
automatically respond to those changes whenever they were made to
jmorecfg.h. The modern reality is that changing RGB_* is no longer
necessary because of the libjpeg-turbo colorspace extensions, and
changing any of the other constants in jsimdcfg.inc can't be done
without making deeper modifications to the SIMD extensions. In general,
we treat RGB_* as a de facto, immutable part of the legacy libpjeg API.
Realistically, since the values of those constants have been the same in
every Un*x distribution released in the past 20-30 years, any software
that uses a system-supplied build of libjpeg must assume that those
constants will have default values.
Furthermore, even if it made sense to auto-generate jsimdcfg.inc, it was
never possible to do so on Windows, so it was always going to be
necessary to manually generate the Windows version of the file whenever
any of the constants changed. This commit introduces a new custom CMake
target called "jsimdcfg" that can be used, on Un*x platforms, to
generate jsimdcfg.inc on demand, although this should only be necessary
when introducing new x86 SIMD instructions or making other deep
modifications, such as SIMD acceleration for 12-bit JPEGs.
For those who may be wondering why we don't do the same thing for
win/jconfig.h.in, it's because performing all of the necessary CMake
checks to populate that file is very slow on Windows.
These were necessary for the first iteration of the feature (see #46),
which provided a different C front end for the SIMD version of the
function. The final version of the feature uses a common C front end
for both SIMD and non-SIMD implementations, so these checks are no
longer necessary.
Closes#231
This commit adds C and SSE2 optimizations for the encode_mcu_AC_first()
function used in progressive Huffman encoding.
The image used for testing can be retrieved from this page:
https://blog.cloudflare.com/doubling-the-speed-of-jpegtran
All timings done on `Intel(R) Core(TM) i7-4870HQ CPU @ 2.50GHz`
clang version is `Apple LLVM version 9.0.0 (clang-900.0.39.2)`
gcc-5 version is `gcc-5 (Homebrew GCC 5.5.0) 5.5.0`
gcc-7 version is `gcc-7 (Homebrew GCC 7.2.0) 7.2.0`
Here are the results in comparison to libjpeg-turbo@293263c using
`time ./jpegtran -outfile /dev/null -progressive -optimise -copy none print_poster_0025.jpg`
C
clang x86_64: +19%
gcc-5 x86_64: +80%
gcc-7 x86_64: +57%
clang i386: +5%
gcc-5 i386: +59%
gcc-7 i386: +51%
SSE2
clang x86_64: +79%
gcc-5 x86_64: +158%
gcc-7 x86_64: +122%
clang i386: +71%
gcc-5 i386: +134%
gcc-7 i386: +135%
Discussion in libjpeg-turbo/libjpeg-turbo#46
This commit adds C and SSE2 optimizations for the encode_mcu_AC_refine()
function used in progressive Huffman encoding.
The image used for testing can be retrieved from this page:
https://blog.cloudflare.com/doubling-the-speed-of-jpegtran
All timings done on `Intel(R) Core(TM) i7-4870HQ CPU @ 2.50GHz`
clang version is `Apple LLVM version 9.0.0 (clang-900.0.39.2)`
gcc-5 version is `gcc-5 (Homebrew GCC 5.5.0) 5.5.0`
gcc-7 version is `gcc-7 (Homebrew GCC 7.2.0) 7.2.0`
Here are the results in comparison to libjpeg-turbo@3c54642 using
`time ./jpegtran -outfile /dev/null -progressive -optimise -copy none print_poster_0025.jpg`
C
clang x86_64: +7%
gcc-5 x86_64: +30%
gcc-7 x86_64: +33%
clang i386: +0%
gcc-5 i386: +24%
gcc-7 i386: +23%
SSE2
clang x86_64: +42%
gcc-5 x86_64: +53%
gcc-7 x86_64: +64%
clang i386: +35%
gcc-5 i386: +46%
gcc-7 i386: +49%
Discussion in libjpeg-turbo/libjpeg-turbo#46
Within the libjpeg API code, it seems to be more the convention than not
to separate the macro name and value by two or more spaces, which
improves general readability. Making this consistent across all of
libjpeg-turbo is less about my individual preferences and more about
making it easy to automatically detect variations from our chosen
formatting convention. I intend to release the script I'm using to
validate this stuff, once it matures and stabilizes a bit.
* Modify the SIMD dispatchers so they guard their usage of getenv() with
the existing NO_GETENV preprocessor definition.
* Introduce a new NO_PUTENV preprocessor definition to guard the
usage of putenv() in the TurboJPEG API library.
This at least puts Windows Store compatibility within the realm of
possibility, although further steps are required.
With rare exceptions ...
- Always separate line continuation characters by one space from
preceding code.
- Always use two-space indentation. Never use tabs.
- Always use K&R-style conditional blocks.
- Always surround operators with spaces, except in raw assembly code.
- Always put a space after, but not before, a comma.
- Never put a space between type casts and variables/function calls.
- Never put a space between the function name and the argument list in
function declarations and prototypes.
- Always surround braces ('{' and '}') with spaces.
- Always surround statements (if, for, else, catch, while, do, switch)
with spaces.
- Always attach pointer symbols ('*' and '**') to the variable or
function name.
- Always precede pointer symbols ('*' and '**') by a space in type
casts.
- Use the MIN() macro from jpegint.h within the libjpeg and TurboJPEG
API libraries (using min() from tjutil.h is still necessary for
TJBench.)
- Where it makes sense (particularly in the TurboJPEG code), put a blank
line after variable declaration blocks.
- Always separate statements in one-liners by two spaces.
The purpose of this was to ease maintenance on my part and also to make
it easier for contributors to figure out how to format patch
submissions. This was admittedly confusing (even to me sometimes) when
we had 3 or 4 different style conventions in the same source tree. The
new convention is more consistent with the formatting of other OSS code
bases.
This commit corrects deviations from the chosen formatting style in the
libjpeg API code and reformats the TurboJPEG API code such that it
conforms to the same standard.
NOTES:
- Although it is no longer necessary for the function name in function
declarations to begin in Column 1 (this was historically necessary
because of the ansi2knr utility, which allowed libjpeg to be built
with non-ANSI compilers), we retain that formatting for the libjpeg
code because it improves readability when using libjpeg's function
attribute macros (GLOBAL(), etc.)
- This reformatting project was accomplished with the help of AStyle and
Uncrustify, although neither was completely up to the task, and thus
a great deal of manual tweaking was required. Note to developers of
code formatting utilities: the libjpeg-turbo code base is an
excellent test bed, because AFAICT, it breaks every single one of the
utilities that are currently available.
- The legacy (MMX, SSE, 3DNow!) assembly code for i386 has been
formatted to match the SSE2 code (refer to
ff5685d5344273df321eb63a005eaae19d2496e3.) I hadn't intended to
bother with this, but the Loongson MMI implementation demonstrated
that there is still academic value to the MMX implementation, as an
algorithmic model for other 64-bit vector implementations. Thus, it
is desirable to improve its readability in the same manner as that of
the SSE2 implementation.
