When attempting to configure an iOS/ARM build with Xcode 7.2 and CMake
2.8.12, I got the following errors:
CMake Error at CMakeLists.txt:560 (add_library):
Attempting to use MACOSX_RPATH without CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG
being set. This could be because you are using a Mac OS X version less
than 10.5 or because CMake's platform configuration is corrupt.
(x 3)
CMake Error at sharedlib/CMakeLists.txt:38 (add_library):
Attempting to use MACOSX_RPATH without CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG
being set. This could be because you are using a Mac OS X version less
than 10.5 or because CMake's platform configuration is corrupt.
(x 3)
Upgrading to CMake 3.x (tried 3.0 and 3.1) got rid of the errors, but
the resulting shared libs still did not use @rpath as expected. Note
also that CMake 3.x (at least the two versions I tested) does not
automatically set the MACOSX_RPATH property as claimed. I could find
nothing in the release notes for later CMake releases to indicate that
either problem has been fixed. What I did find was this little nugget
of code in the Darwin platform module:
f6b93fbf3a/Modules/Platform/Darwin.cmake (L33-L36)
This sets CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG="-Wl,-rpath," only if you
are running OS X 10.5 or later. It makes no such check for iOS, perhaps
because shared libraries aren't much of a thing with iOS apps. In any
event, this commit simply sets CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG if it
isn't set already, and that fixes all of the aforementioned problems.
- Travis doesn't set the $encrypted_* variables for PRs, so disable GPG
signing when building a PR (artifacts aren't deployed for PRs anyhow,
and even if they were, I wouldn't want them to be signed, as they may
contain unvetted code.)
- Take advantage of the new -d option in buildljt, which allows for
building from an existing Git clone directory. This eliminates the need
to rename and restore .git/shallow, allows the official build scripts to
work properly when building PRs, and prevents 'git clone' being invoked
twice in CI builds.
Refer to #217
These files are potentially useful to MinGW users, since MSYS2 MinGW
environments have a man command by default and provide an easy way to
install pkg-config.
Closes#223
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.
Broken by previous commit. Although turbojpeg.c no longer needs
tjutil.h on Un*x, it still needs to include that file on Windows in
order to use snprintf() and strcasecmp() (which, on Windows, are macros
that wrap _snprintf_s() and stricmp().)
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.
+ "JSIMD_ARM_NEON" = "JSIMD_NEON"
+ "JSIMD_MIPS_DSPR2" = "JSIMD_DSPR2"
+ "*_mips_dspr2" = "*_dspr2"
It's obvious that "NEON" refers to Arm and "DSPr2" refers to MIPS, and
this naming convention is consistent with the other SIMD extensions.
Newer versions of CMake (known to be the case with 3.7.x and 3.10.x)
fail to add a space between CMAKE_C_FLAGS and CMAKE_ASM_FLAGS, which
causes the build to fail when using the official build procedure.
Closes#216
Referring to https://docs.microsoft.com/en-US/cpp/build/stack-usage:
"All memory beyond the current address of RSP is considered volatile:
The OS, or a debugger, may overwrite this memory during a user debug
session, or an interrupt handler. Thus, RSP must always be set before
attempting to read or write values to a stack frame."
Basically, if-- under extremely rare circumstances-- a context swap were
to occur between saving the values of xmm8-xmm11 and setting the new
value of rsp, the O/S might not preserve that area of the stack. In
general, libjpeg-turbo should not be using xmm8-xmm11 before or after
the call to jsimd_huff_encode_one_block_sse2(), so this is probably a
non-issue, but it's still a good idea to fix it.
Based on
ff7d2030dd
NDK r16b moved some things around, so modify the Android build recipes
to take that into account while preserving compatibility with previous
NDK releases.
NOTE: the GCC 4.9 NDK toolchain is deprecated, so we will need to
develop new Android build recipes for libjpeg-turbo 1.6 that use the
Clang toolchain.
Closes#196
If jpeg_skip_scanlines() is used to skip to the end of a single-scan
image, then we need to change the library state such that subsequent
calls to jpeg_consume_input() will return JPEG_REACHED_EOI rather than
JPEG_SUSPENDED. (NOTE: not necessary for multi-scan images, since the
scans are processed prior to any call to jpeg_skip_scanlines().)
Unless I miss my guess, using jpeg_skip_scanlines() in this manner
will prevent any markers at the end of the JPEG image from being
read, but I don't think there is any way around that without actually
reading the data, which would defeat the purpose of
jpeg_skip_scanlines().
Fixes#194
Refer to travis-ci/travis-ci#8552. This was supposed to be fixed on
November 15, then on November 28. Travis blew through both deadlines,
so I have no confidence that the issue will be fixed as promised in a
timely manner. Adding 'brew update' to .travis.yml slows the OS X
build, but there is no choice at the moment.
Loading RGB image files into a grayscale buffer isn't a particularly
useful feature, given that libjpeg-turbo can perform this conversion
much more optimally (with SIMD acceleration on some platforms) during
the compression process. Also, the RGB2GRAY() macro was not producing
deterministic cross-platform results because of variations in the
round-off behavior of various floating point implementations, so
`tjunittest -bmp` was failing in i386 builds.
Also, set the red/green/blue offsets for TJPF_GRAY to -1 rather than 0.
It was undefined behavior for an application to use those arrays/methods
with TJPF_GRAY anyhow, and this makes it easier for applications to
programmatically detect whether a given pixel format has red, green, and
blue components.
The main justification for this is to provide new libjpeg-turbo users
with a quick & easy way of developing a complete JPEG
compression/decompression program without requiring them to build
libjpeg-turbo from source (which was necessary in order to use the
project-private bmp API) or to use external libraries. These new
functions build upon significant enhancements to rdbmp.c, wrbmp.c,
rdppm.c, and wrppm.c which allow those engines to convert directly
between the native pixel format of the file and a pixel format
("colorspace" in libjpeg parlance) specified by the calling program.
rdbmp.c and wrbmp.c have also been modified such that the calling
program can choose to read or write image rows in the native (bottom-up)
order of the file format, thus eliminating the need to use an inversion
array. tjLoadImage() and tjSaveImage() leverage these new underlying
features in order to significantly improve upon the performance of the
old bmp API.
Because these new functions cannot work without the libjpeg-turbo
colorspace extensions, the libjpeg-compatible code in turbojpeg.c has
been removed. That code was only there to serve as an example of how
to use the TurboJPEG API on top of libjpeg, but more specific, buildable
examples now exist in the https://github.com/libjpeg-turbo/ijg
repository.
tjbenchtest and its Java derivatives are useful for rooting out hidden
problems with the more esoteric TJBench and TurboJPEG features. For
instance, on Windows, running tjbenchtest uncovered
5fce2e9421.
This commit also causes tjbenchtest and tjbenchtest.java to append -yuv
and -alloc to their log file names, depending on the arguments passed,
and it causes the build system to clean up those log files when the
'testclean' target is built.
+ clean up log files when 'make testclean' is invoked
+ fix 'tjbenchtest -yuv -alloc'
+ fix tjexampletest so that it creates images under /tmp
+ clean up tjexampletest
The program crashed when a JPEG image was passed on the command line,
because we were mixing our metaphors vis-a-vis malloc()/free() and
tjAlloc()/tjFree() (malloc()/free() uses the tjbench.exe heap,
whereas tjAlloc()/tjFree() uses the turbojpeg.dll heap.)
- Referring to 073b0e88a1 and #185, the
reason why BMP and RLE didn't (and won't) work with partial image
decompression is that the output engines for both formats maintain a
whole-image buffer, which is used to reverse the order of scanlines.
However, it was straightforward to add -crop support for GIF and
Targa, which is useful for testing partial image decompression along
with color quantization.
- Such testing reproduced a bug reported by Mozilla (refer to PR #182)
whereby jpeg_skip_scanlines() would segfault if color quantization was
enabled. To fix this issue, read_and_discard_scanlines() now sets up
a dummy quantize function in the same manner that it sets up a dummy
color conversion function.
Closes#182
... and document that only PPM/PGM output images are supported with the
-crop option for the moment.
I investigated the possibility of supporting -crop with -bmp, but even
after resetting the buffer dimensions, I still kept getting virtual
array access errors. It seems that doing this the "right way" would
require creating a re-initialization function for each image format's
destination manager. I'm disinclined to do that right now, given that
this feature was Google's baby (developed as a prerequisite for
including libjpeg-turbo in Android), and the -crop option in djpeg is
intended only as an example of how to use the partial image
decompression API. Real-world applications would need to handle this
in their own destination managers.
It would probably be possible to make this work with Targa by employing
a similar hack to the one we used with PPM, but Targa isn't popular
enough to bother.
Fixes#185
Setting _libdir to CMAKE_INSTALL_FULL_LIBDIR only works when doing an
in-tree RPM build. SRPMs are architecture-agnostic, so the spec needs
to compute_libdir at the time the SRPM is rebuilt, not at the time it
is generated.
This is a regression introduced when implementing the new CMake-based
cross-platform build system.
The version of RPM on RHEL 5 and older platforms defines _libdir
as %{_exec_prefix}/%{_lib}, so defining _lib in the spec file redefined
_libdir. However, newer versions of RPM (probably >= 4.6, since that
was the version that introduced the ISA macros) define _libdir as either
%{_prefix}/lib or %{_prefix}/lib64. Thus, we need to explicitly
override _libdir in our spec file.
It is necessary for the C code to be aware of the machine's endianness,
which is why the TurboJPEG Java wrapper sets a different pixel format
for integer BufferedImages depending on ByteOrder.nativeOrder().
However, it isn't necessary to handle endianness in pure Java code such
as TJUnitTest (d'oh!) This was a product of porting the C version of
TJUnitTest too literally, and of insufficient testing (historically,
the big endian systems I had available for testing didn't have Java.)
planes == null is a valid argument to setBuf() if alloc == true, so we
need to make sure that planes is non-null before validating its length.
We also need to allocate one dimension of the planes array if it's null.
Fixes#168
Previously, -stoponwarning only had an effect on the underlying
TurboJPEG C functions, but TJBench still aborted if a non-fatal error
occurred. This commit modifies the C version of TJBench such that it
always recovers from a non-fatal error unless -stoponwarning is
specified. Furthermore, the benchmark stores the details of the last
non-fatal error and does not print any subsequent non-fatal error
messages unless they differ from the last one.
Due to limitations in the Java API (specifically, the fact that it
cannot communicate errors, fatal or otherwise, to the calling program
without throwing a TJException), it was only possible to make
decompression operations fully recoverable within TJBench. With other
operations, -stoponwarning still has an effect on the underlying C
library but has no effect at the Java level.
The Java API documentation has been amended to reflect that only certain
methods are truly recoverable, regardless of the state of
TJ.FLAG_STOPONWARNING.
Allow progressive entropy coding to be enabled on a
transform-by-transform basis, and implement a new transform option for
disabling the copying of markers.
Closes#153
If the source image for a transform operation has a lot of EXIF or ICC
data embedded in it, then it may cause the output image size to exceed
the worst-case size returned by tjBufSize() (because tjTransform()
transfers all markers to the output image.) This is only a problem if
TJFLAG_NOREALLOC is passed to the function. Since the TurboJPEG C API
doesn't require the destination image size to be set in this case, it
makes the documented assumption that the calling program has allocated
the destination buffer to exactly the size returned by tjBufSize().
Changing this assumption would change the API behavior and necessitate
a new function name (tjTransform2().) At the moment, it's easier to
just document this as a known issue, since it's easy to work around in
the C API.
The Java API is unfortunately a different story, since it must always
use TJFLAG_NOREALLOC (because, when using the TurboJPEG Java API, all
buffers are allocated on the Java heap, and thus they can't be
reallocated by the C code.) There is no easy way to work around this
without changing the C API as discussed above, because if the source
image contains large amounts of marker data, it's virtually impossible
to determine how big the output image will be.
Given that libjpeg-turbo can often process hundreds of megapixels/second
on modern hardware, the default of one warmup iteration was essentially
meaningless. Furthermore, the -warmup option was a bit clunky, since
it required some foreknowledge of how fast the benchmarks were going to
execute.
This commit introduces a 1-second warmup interval for each benchmark by
default, and the -warmup option has been retasked to control the length
of that interval.
- Provide a new C API function and TJException method that allows
calling programs to query the severity of a compression/decompression/
transform error.
- Provide a new flag that instructs the library to immediately stop
compressing/decompressing/transforming if a warning is encountered.
Fixes#151
Introduce a new C API function (tjGetErrorStr2()) that can be used to
retrieve compression/decompression/transform error messages in a
thread-safe (i.e. instance-specific) manner. Retrieving error messages
from global functions is still thread-unsafe.
Addresses a concern expressed in #151.
Embedded ICC profiles can cause the size of a JPEG file to exceed the
size returned by tjBufSize() (which is really meant to be used for
compression anyhow, not for decompression), and this was causing a
segfault (C) or an ArrayIndexOutOfBoundsException (Java) when
decompressing such files with TJBench. This commit modifies the
benchmark such that, when tiled decompression is disabled, it re-uses
the source buffer as the primary JPEG buffer.
The Travis xcode7.3 image now apparently includes GnuPG 1.4.x by
default, so use it instead of installing GnuPG 2. Using GnuPG 2.1.x,
the default version in Homebrew as of this writing, is problematic for
this reason:
https://wiki.archlinux.org/index.php/GnuPG#Unattended_passphrase
This re-introduces a feature of the obsolete system-specific libjpeg
memory managers-- namely the ability to limit the amount of main memory
used by the library during decompression or multi-pass compression.
This is mainly beneficial for two reasons:
- Works around a 2 GB limit in libFuzzer
- Allows security-sensitive applications to set a memory limit for the
JPEG decoder so as to work around the progressive JPEG exploit
(LJT-01-004) described here:
http://www.libjpeg-turbo.org/pmwiki/uploads/About/TwoIssueswiththeJPEGStandard.pdf
This commit also removes obsolete documentation regarding the MS-DOS
memory manager (which itself was removed long ago) and changes the
documentation of the -maxmemory switch and JPEGMEM environment variable
to reflect the fact that backing stores are never used in libjpeg-turbo.
Inspired by:
066fee2e7dCloses#143
Something changed in the CI build environment, and our previous trick of
setting the Git URL to file://c:/projects/libjpeg-turbo no longer works.
Using cygpath to translate the Windows path to a MinGW-friendly format
is a better solution anyhow.
Referring to https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=746,
it seems that the values of local buffer pointers in TurboJPEG API
functions aren't always preserved if longjmp() returns control to a
point prior to the allocation of the local buffers. This is known to
be an issue with GCC 4.x and clang with -O1 and higher optimization
levels but not with GCC 5.x and later. It is unknown why GCC 5.x and
6.x do not suffer from the issue, but possibly the local buffer pointers
are not allocated on the stack when using those more recent compilers.
In any case, this commit modifies the TurboJPEG API library code such
that the jump buffer is always updated after any local buffer pointers
are allocated but before any subsequent libjpeg API functions are
called.
This commit does the following:
-- Merges the two glueware functions (read_icc_profile() and
write_icc_profile()) from iccjpeg.c, which is contained in downstream
projects such as LCMS, Ghostscript, Mozilla, etc. These functions were
originally intended for inclusion in libjpeg, but Tom Lane left the IJG
before that could be accomplished. Since then, programs and libraries
that needed to embed/extract ICC profiles in JPEG files had to include
their own local copy of iccjpeg.c, which is suboptimal.
-- The new functions were prefixed with jpeg_ and split into separate
files for the compressor and decompressor, per the existing libjpeg
coding standards.
-- jpeg_write_icc_profile() was made slightly more fault-tolerant.
It will now trigger a libjpeg error if it is called before
jpeg_start_compress() or if it is passed NULL arguments.
-- jpeg_read_icc_profile() was made slightly more fault-tolerant.
It will now trigger a libjpeg error if it is called before
jpeg_read_header() or if it is passed NULL arguments. It will also
now trigger libjpeg warnings if the ICC profile data is corrupt.
-- The code comments have been wordsmithed.
-- Note that the one-line setup_read_icc_profile() function was not
included. Instead, libjpeg.txt now documents the need to call
jpeg_save_markers(cinfo, JPEG_APP0 + 2, 0xFFFF) prior to calling
jpeg_read_header(), if jpeg_read_icc_profile() is to be used.
-- Adds documentation for the new functions to libjpeg.txt.
-- Adds an -icc switch to cjpeg and jpegtran that allows those programs
to embed an ICC profile in the JPEG files they generate.
-- Adds an -icc switch to djpeg that allows that program to extract an
ICC profile from a JPEG file while decompressing.
-- Adds appropriate unit tests for all of the above.
-- Bumps the SO_AGE of the libjpeg API library to indicate the presence
of new API functions.
Note that the licensing information was obtained from:
https://github.com/mm2/Little-CMS/issues/37#issuecomment-66450180
... even if using libjpeg v6b emulation. Previously
adjust_exif_parameters() was only called with libjpeg v7/v8 emulation,
but due to a bug (which this commit also fixes), it only worked properly
with libjpeg v8 emulation.
The JPEG_LIB_VERSION #ifdef in jtransform_adjust_parameters() was
incorrect, which caused a "Bogus virtual array access" error when
attempting to use the lossless crop feature.
Introduced in c04bd3cc97.
This also adds libjpeg v7 API/ABI emulation to the Travis CI tests.
LICENSE.md is included in the binary distributions as well, so it
doesn't make much sense to refer to license headers in source files that
aren't necessarily going to be there.
The whole point of `make tarball` is to make it easy for users to create
a binary distribution of libjpeg-turbo on platforms that aren't
supported by our official build system, so requiring root permissions
somewhat defeated that purpose. Intead, the script now attempts to
detect whether the system has GNU tar or a recent version of BSD tar
that supports setting the ownership of the files in the tarball.
Although there is little chance that we will ever have a package
conflict on OS X, the convention from our Linux packages is to use the
package name, not the project name, for the name of the documentation
directory.
These improvements enable build systems to use GNUInstallDirs to define
custom directory variables.
- The set_dir() macro was renamed to GNUInstallDirs_set_install_dir(),
in keeping with the module's established macro naming convention.
- Rather than detecting whether the prefix has changed, the new
GNUInstallDirs_set_install_dir() macro instead examines whether the
default for the variable in question has changed. This allows for
more flexibility, since build systems may decide to change the
defaults based on factors other than the prefix. It also enables the
macro to work properly outside of the module.
- The module now performs directory variable substitution within the
body of GNUInstallDirs_get_absolute_install_dir().
- The JAVADIR variable is no longer included in GNUInstallDirs. That
directory is not part of the GNU spec, and it turns out that various
operating systems use different conventions for the location of Java
classes. Instead, the variable is now implemented in our build
system as a demonstration of the aforementioned GNUInstallDirs
enhancements.
- GNUInstallDirs: any directory variable can now reference any other
directory variable by including its name in angle brackets (<>).
- Changed the documentation of the directory variables in BUILDING.md
accordingly. This commit also includes some formatting tweaks to
that section (using boldface for directory names, as is our
convention.)
- Changed the package scripts such that they use
CMAKE_INSTALL_DATAROOTDIR rather than CMAKE_INSTALL_DATADIR.
- We no longer override the install dir. defaults on Windows unless
performing an official build. It may be useful, for instance, to
use the GNU defaults when installing into an MSYS environment.
It isn't actually necessary to specify `CMAKE_INSTALL_DEFAULT_MANDIR`
for our official build. Because `CMAKE_INSTALL_DEFAULT_DATAROOTDIR` is
blank for the official build, the default of "<DATAROOTDIR>/man" will
resolve to "man".
For the same reason, this commit changes the specification of
`CMAKE_INSTALL_DEFAULT_DOCDIR` and `CMAKE_INSTALL_DEFAULT_JAVADIR` in
the official build to be dependent on the data root directory (mainly to
make it obvious what we're doing.)
This commit also tweaks the example CMake command line in the directory
variable documentation so that it shows the correct location of the
CMake argument.
YASM requires a debug format to be specified with -g. Currently the
only combination that I can make work at all is DWARF-2/ELF (YASM
doesn't support Mach-O debugging at all, and its support for CV8/MSVC
and MinGW/DWARF-2 appears to be broken), so debugging is only enabled
automatically for ELF at the moment. For other formats, we don't
specify -g at all, which is how the old build system behaved.
Fixes#125, Closes#126
This builds upon the existing GNUInstallDirs module in CMake but adds
the following features to that module:
- The ability to override the defaults for each install directory
through a new set of variables (`CMAKE_INSTALL_DEFAULT_*DIR`).
Before operating system vendors began shipping libjpeg-turbo, it was
meant to be a run-time drop-in replacement for the system's
distribution of libjpeg, so it has traditionally installed itself
under /opt/libjpeg-turbo on Un*x systems by default. On Windows, it
has traditionally installed itself under %SystemDrive%\libjpeg-turbo*,
which is not uncommon behavior for open source libraries (open source
SDKs tend to install outside of the Program Files directory so as to
avoid spaces in the directory name.) At least in the case of Un*x,
the install directory behavior is based somewhat on the Solaris
standard, which requires all non-O/S packages to install their files
under /opt/{package_name}. I adopted that standard for VirtualGL and
TurboVNC while working at Sun, because it allowed those packages to be
located under the same directory on all platforms. I adopted it for
libjpeg-turbo because it ensured that our files would never conflict
with the system's version of libjpeg. Even though many Un*x
distributions ship libjpeg-turbo these days, not all of them ship the
TurboJPEG API library or the Java classes or even the latest version
of the libjpeg API library, so there are still many cases in which it
is desirable to install a separate version of libjpeg-turbo than the
one installed by the system. Furthermore, installing the files under
/opt mimics the directory structure of our official binary packages,
and it makes it very easy to uninstall libjpeg-turbo.
For these reasons, our build system needs to be able to use
non-GNU-compliant defaults for each install directory if
`CMAKE_INSTALL_PREFIX` is set to the default value.
- For each directory variable, the module now detects changes to
`CMAKE_INSTALL_PREFIX` and changes the directory variable accordingly,
if the variable has not been changed by the user.
This makes it easy to switch between our "official" directory
structure and the GNU-compliant directory structure "on the fly"
simply by changing `CMAKE_INSTALL_PREFIX`. Also, this new mechanism
eliminated the need for the crufty mechanism that previously did the
same thing just for the library directory variable.
How it should work:
- If a dir variable is unset, then the module will set an internal
property indicating that the dir variable was initialized to its
default value.
- If the dir variable ever diverges from its default value, then the
internal property is cleared, and it cannot be set again without
unsetting the dir variable.
- If the install prefix changes, and if the internal property
indicates that the dir variable is still set to its default value,
and if the dir variable's value is not being manually changed at the
same time that the install prefix is being changed, then the dir
variable's value is automatically changed to the new default value
for that variable (as determined by the new install prefix.)
- The directory variables are now always cached, regardless of whether
they were set on the command line or not. This ensures that they can
easily be examined and modified after being set, regardless of how they
were set.
This was made possible by the introduction of the aforementioned
`CMAKE_INSTALL_DEFAULT_*DIR` variables.
- Improved directory variable documentation (based on descriptions at
https://www.gnu.org/prep/standards/html_node/Directory-Variables.html)
- The module now allows "<DATAROOTDIR>" to be used as a placeholder in
relative directory variables.
It is replaced "on the fly" with the actual path of
`CMAKE_INSTALL_DATAROOTDIR`.
This should more closely mimic the behavior of the old autotools build
system while retaining our customizations to it, and it should retain
the behavior of the old CMake build system.
Closes#124
Strict C89-conformant compilers don't support the "inline" keyword, but
most of them support "__inline__", and that keyword can be used with the
always_inline atribute as well. This commit also removes duplicate code
by using a foreach() loop to test the various keywords.
This is a subtle point, but AC_C_INLINE defines "inline" to be either
"inline", "__inline__", or "__inline". The subsequent test for
"inline __attribute__((always_inline))" uses this definition. The
attribute is irrespective of the inline keyword, so whereas
"__inline__ __attribute__((always_inline))" works under C89,
"inline __attribute__((always_inline))" doesn't, and defining INLINE to
the latter causes the build to fail. The easiest way around this is
simply to define "inline" ahead of "INLINE" in jconfigint.h,
which causes the inline keyword detected by AC_C_INLINE to modify the
INLINE macro if necessary.
+ advertise that full AltiVec SIMD acceleration is now available on
OpenBSD.
The relevant compilers probably all support C99 or GNU's variation of
C90 that allows variables to be declared anywhere, but our policy is to
conform to the C90 standard, if for no other reason than that it
improves code readability.
Fixes:
rdppm.c:257:14: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
if (temp > maxval)
~~~~ ^ ~~~~~~
rdppm.c:284:14: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
if (temp > maxval)
~~~~ ^ ~~~~~~
rdppm.c:289:14: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
if (temp > maxval)
~~~~ ^ ~~~~~~
rdppm.c:294:14: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
if (temp > maxval)
- Replace CMAKE_SOURCE_DIR with CMAKE_CURRENT_SOURCE_DIR
- Replace CMAKE_BINARY_DIR with CMAKE_CURRENT_BINARY_DIR
- Don't use "libjpeg-turbo" in any of the package system filenames
(because CMAKE_PROJECT_NAME will not be the same if building LJT as
a submodule.)
Closes#122
The xcode7.2 image is verfallen, verlumpt, verblunget, verkackt
This also ensures that the build scripts are checked out from a
branch matching the libjpeg-turbo repository branch (not strictly
necessary when building from master, but it keeps the code in sync with
dev.)
The previous hack (adding ${CMAKE_ASM_COMPILER} to CMAKE_ASM_FLAGS)
didn't work in all cases, because more recent versions of CMake place
the includes ahead of the flags (which meant that the real assembler
wasn't the first argument to gas-preprocessor.pl.)
CMAKE_INSTALL_RPATH has to be set before the targets are defined (oops.)
This also explicitly turns on MACOSX_RPATH for the shared libraries
(which is the default with newer versions of CMake but not with 2.8.x.)
The old autotools/libtool build system hard-coded the install name
directory of the OS X shared libraries to libdir, which meant that any
executable that linked against those libraries would also be hard-coded
to look for the libjpeg-turbo libraries in that directory. @rpath makes
the OS X version of libjpeg-turbo behave like the Linux version, in the
sense that the executables under /opt/libjpeg-turbo/bin will
automatically pick up the libraries under /opt/libjpeg-turbo/lib* by
default, but other executables won't unless they are linked with -rpath.
AppVeyor already has MinGW32 and MinGW64 flavors of GCC 5.3.0
installed under MSYS2, so there is no need to install our own builds of
MinGW. MinGW-builds is no longer an active project, and we were getting
occasional timeouts while wgetting those files from SourceForge.
Furthermore, GCC 5.3.0 should produce faster code than GCC 4.8.1.
Updated out-of-date information, wordsmithed and clarified many
sections, and generally cleaned up the build recipes (including a
complete overhaul of the iOS recipes.)
Regression caused by f9134384b7
This commit also makes the "testclean" target clean up the 4:1:1 test
images. This was implemented in the autotools build system in
1f3635c496 but was left out of the CMake
build system due to an oversight.
This has the following advantages:
-- It doesn't require checking a private SSH key into the repository.
(With SourceForge, an SSH key is the "keys to the kingdom".)
-- If the S3 key is compromised, it is very easy to revoke it and
generate a new one.
-- The S3 bucket is isolated, so even if it becomes compromised, then
the damage that one could do is limited.
-- It's much easier to manage files through S3's web interface than
through SourceForge.
-- The files are served via HTTPS.
-- Travis fully supports S3 as a deployment target, so this simplifies
.travis.yml somewhat.
Since we're still deploying our Linux/macOS CI artifacts to a web server
(specifically SourceForge Project Web Services) that doesn't support
HTTPS, it's a good idea to sign them. But since the private key has to
be checked into the repository, we use a different key for signing the
pre-releases (per project policy, the private signing keys for our
release binaries are never made available on any public server.)
Previously, simd/CMakeLists.txt was hard-coded to use NASM, and it was
necessary to override the NASM variable in order to use YASM. This
commit changes the behavior such that NASM is still preferred, but YASM
will be used if it is in the PATH and NASM isn't available. This brings
the actual behavior in line with the behavior described in BUILDING.md.
Based on
b0799a1598Closes#107
-- Use trusty for SIMD builds. Ubuntu 12.04 is still using NASM 2.09.x,
which isn't new enough to support AVX2.
-- Add a special test for the SSE2 code path, since it is no longer the
default.
Pass the actual repository and branch that Travis is using into the
builtljt script, so the official builds it generates will come from
the same code base as the other tested builds.
- Introduce a new FLOATTEST value ("387") on Un*x systems that will
compare the floating point DCT/IDCT algorithms against the expected
results from the C algorithms when built using 32-bit code and
-mfpmath=387.
- Extend the Windows regression tests so that they work properly when
building libjpeg-turbo with 32-bit code and without SIMD, using either
Visual C++ (tested with 2008, 2010, 2015) or MinGW.
Based on
98a5a9dc89
with wordsmithing by DRC.
In the AArch64 ABI, as in many others, it's forbidden to read/store data
below the stack pointer. Some SIMD functions were doing just that
(stack pointer misuse) when trying to preserve callee-saved registers,
and this resulted in those registers being restored with incorrect
contents under certain circumstances.
This patch fixes that behavior, and callee-saved registers are now
stored above the stack pointer throughout the function call. The patch
also removes register saving in places where it is unnecessary for this
ABI, or it makes use of unused scratch regiters instead of callee-saved
registers.
Fixes#97. Closes#101.
Refer also to https://bugzilla.redhat.com/show_bug.cgi?id=1368569
The last iDevice to require ARMv6 was the iPhone 3G, which required iOS
4.2.1 or older. Our binaries have always required iOS 4.3 or newer,
so I'm not sure if the ARMv6 fork of our binaries was ever useful to
begin with. In any case, if it ever was useful, it no longer is. Fat
binaries can still be generated with ARMv6 support by invoking
{build_directory}/pkgscripts/makemacpkg manually.
Reported by Clang UBSan (refer to
https://bugzilla.mozilla.org/show_bug.cgi?id=1301252 for test image.)
This appears to be a legitimate bug introduced by
3ab68cf563. Any component array, such
as first_MCU_col and last_MCU_col, should always be able to accommodate
MAX_COMPONENTS values. The aforementioned test image had 8 components,
which was not enough to make the out-of-bounds write bust out of the
jpeg_decomp_master struct (and fortunately the memory after last_MCU_col
is an integer used as a boolean, so stomping on it will do nothing other
than change the decoder state.) I crafted another special image that
has 10 components (the maximum allowable), but that was apparently not
enough to bust out of the allocated memory, either. Thus, it is
posited that the security threat posed by this bug is either extremely
minimal or non-existent.
When attempting to decode a malformed JPEG image (refer to
https://bugzilla.mozilla.org/show_bug.cgi?id=1295044) with dimensions
61472 x 32800, the maximum_space variable within the
realize_virt_arrays() function will exceed the maximum value of a 32-bit
integer and will wrap around. The memory manager subsequently fails
with an "Insufficient memory" error (case 4, in alloc_large()), so this
commit simply causes that error to be triggered earlier, before UBSan
has a chance to complain.
Note that this issue did not ever represent an exploitable security
threat, because the POSIX-based memory manager that we use doesn't ever
do anything meaningful with the value of maximum_space.
jpeg_mem_available() simply sets avail_mem = maximum_space, so the
subsequent behavior of the memory manager is the same regardless of
whether maximum_space is correct or not. This commit simply removes a
UBSan warning in order to make it easier to detect actual security
issues.
Normally, 4:2:2 JPEGs have horizontal x vertical luminance,chrominance
sampling factors of 2x1,1x1, and 4:4:0 JPEGs have horizontal x vertical
luminance,chrominance sampling factors of 1x2,1x1. However, it is
technically legal to create 4:2:2 JPEGs with sampling factors of
2x2,1x2 and 4:4:0 JPEGs with sampling factors of 2x2,2x1, since the
sums of the products of those sampling factors (2x2 + 1x2 + 1x2 and
2x2 + 2x1 + 2x1) are still <= 10. The libjpeg API correctly decodes
such images, so the TurboJPEG API should as well.
Fixes#92
Currently, this only affects ARM, since it is the only platform that
accelerates YCbCr-to-RGB conversion but not merged upsampling. Even if
"plain" upsampling isn't accelerated, the combination of accelerated
color conversion + unaccelerated plain upsampling is still faster than
the unaccelerated merged upsampling algorithms.
Closes#81
This allows fancy upsampling to be used when decompressing 4:2:2 images
that have been losslessly rotated or transposed.
(docs and comments added by DRC)
Based on f63aca945dCloses#89
cpuid tells us whether the O/S uses extended state management via
XSAVE/XRSTOR, but we have to call xgetbv to verify that it is using
XSAVE/XRSTOR to manage the state of XMM/YMM registers.
In the AArch64 ABI, the high (unused) DWORD of a 32-bit argument's
register is undefined, so it was incorrect to use 64-bit
instructions to transfer a JDIMENSION argument in the 64-bit NEON SIMD
functions. The code worked thus far only because the existing compiler
optimizers weren't smart enough to do anything else with the register in
question, so the upper 32 bits happened to be all zeroes.
The latest builds of Clang/LLVM have a smarter optimizer, and under
certain circumstances, it will attempt to load-combine adjacent 32-bit
integers from one of the libjpeg structures into a single 64-bit integer
and pass that 64-bit integer as a 32-bit argument to one of the SIMD
functions (which is allowed by the ABI, since the upper 32 bits of the
32-bit argument's register are undefined.) This caused the
libjpeg-turbo regression tests to crash.
This patch tries to use the Wn registers whenever possible. Otherwise,
it uses a zero-extend instruction to avoid using the upper 32 bits of
the 64-bit registers, which are not guaranteed to be valid for 32-bit
arguments.
Based on 1fbae13021Closes#91. Refer also to android-ndk/ndk#110 and
https://llvm.org/bugs/show_bug.cgi?id=28393
This fixes crashes that would occur when attempting to use
libjpeg-turbo's AVX2 extensions on older O/S's (such as Windows XP or
RHEL 5.) Even if the CPU supports AVX2, the O/S has to also support
saving/restoring YMM registers when switching contexts.
This eliminates "illegal instruction" errors when running libjpeg-turbo
under Linux on PowerPC chips that lack AltiVec support (e.g. the old
7XX/G3 models but also the newer e5500 series.)
The JSIMD_FORCE* environment variables previously meant "force the use
of this instruction set if it is available but others are available as
well", but that did nothing on ARM platforms, since there is only ever
one instruction set available. Since the ARM and MIPS CPU feature
detection code is less than bulletproof, and since there is only one
SIMD instruction set (currently) supported on those platforms, it makes
sense for the JSIMD_FORCE* environment variables on those platforms to
actually force the use of the SIMD instruction set, thus bypassing the
CPU feature detection code.
This addresses a concern raised in #88 whereby parsing /proc/cpuinfo
didn't work within a QEMU environment. This at least provides a
workaround, allowing users to force-enable or force-disable SIMD
instructions for ARM and MIPS builds of libjpeg-turbo.
This commit adds back instructive comments in the image-space
algorithms, similar to those in the SSE2 code. These comments make it
easier to follow the flow of data through the algorithms.
Expand collect_args/uncollect_args macros so that the number of
arguments can be specified. This prevents unnecessary push and mov
instructions.
NOTE: On Windows, the push/pop of xmm6 and xmm7 had to be moved to the
other end of the macro to ensure that rsp is aligned on a 16-byte
boundary.
28d1a1300c introduced the line
"nasm.exe should be in your PATH". This commit corrects an oversight in
8f1c0a681c /
e5091f2cf3 whereby this line should have
been extended to include yasm.exe.
The IJG convention is to format copyright notices as:
Copyright (C) YYYY, Owner.
We try to maintain this convention for any code that is part of the
libjpeg API library (with the exception of preserving the copyright
notices from Cendio's code verbatim, since those predate
libjpeg-turbo.)
Note that the phrase "All Rights Reserved" is no longer necessary, since
all Buenos Aires Convention signatories signed onto the Berne Convention
in 2000. However, our convention is to retain this phrase for any files
that have a self-contained copyright header but to leave it off of any
files that refer to another file for conditions of distribution and use.
For instance, all of the non-SIMD files in the libjpeg API library refer
to README.ijg, and the copyright message in that file contains "All
Rights Reserved", so it is unnecessary to add it to the individual
files.
The TurboJPEG code retains my preferred formatting convention for
copyright notices, which is based on that of VirtualGL (where the
TurboJPEG API originated.)
Calling jpeg_stdio_dest() followed by jpeg_mem_dest(), or jpeg_mem_src()
followed by jpeg_stdio_src(), is dangerous, because the existing opaque
structure would not be big enough to accommodate the new source/dest
manager. This issue was non-obvious to libjpeg-turbo consumers, since
it was only documented in code comments. Furthermore, the issue could
also occur if the source/dest manager was allocated by the calling
program, but it was not allocated with enough space to accommodate the
opaque stdio or memory source/dest manager structs. The safest thing to
do is to throw an error if one of these functions is called when there
is already a source/dest manager assigned to the object and it was
allocated elsewhere.
Closes#78, #79
The jpeg-7/jpeg-8 APIs/ABIs require arithmetic coding, and the jpeg-8
API/ABI requires the memory source/destination manager, so this commit
causes the build system to ignore --with-arith-enc/--without-arith-enc
and --with-arith-dec/--without-arith-dec (and the equivalent CMake
variables-- WITH_ARITH_ENC and WITH_ARITH_DEC) when v7/v8 API/ABI
emulation is enabled. Furthermore, the CMake build system now ignores
WITH_MEM_SRCDST whenever WITH_JPEG8 is specified (the autotools build
system already did that.)
GCC does support UAL syntax (strbeq) if the ".syntax unified" directive
is supplied. This directive is supported by all versions of GCC and
clang going back to 2003, so it should not create any backward
compatibility issues.
Based on 1264349e2fCloses#76
If wmic.exe wasn't available, then CMakeLists.txt would call
"cmd /C date /T" and parse the result in order to set the BUILD
variable. However, the parser assumed that the date was in MM/DD/YYYY
format, which is not generally the case unless the user's locale is U.S.
English with the default region/language settings for that locale.
This commit modifies CMakeLists.txt such that it uses the
string(TIMESTAMP) function available in CMake 2.8.11 and later to set
the BUILD variable, thus eliminating the need to use wmic.exe or any
other platform-specific hack.
This commit also modifies the build instructions to remove any reference
to CMake 2.6 (which hasn't been supported by our build system since
libjpeg-turbo 1.3.x.)
Closes#74
At one time, it was possible to use CMake to build under Cygwin, but
that hasn't worked since 1.4.1 (due to the Huffman codec changes that
now require SIZEOF_SIZE_T to be defined for non-WIN32 platforms) and may
have even been broken before that. Originally, we used the "date"
command under MSYS in order to obtain the default build number, but that
was rendered unnecessary by 5e3bb3e9 (v1.3 beta.) 9fe22dac (1.4 beta)
further modified CMakeLists.txt so that the "date" command was only used
on Cygwin, but for unexplained reasons, that commit also applied the
(now vestigial) code to all non-WIN32 platforms. This prevented
CMakeLists.txt from displaying an error if someone attempted to use the
CMake build system on Un*x platforms, and that may have been behind the
flurry of pull requests and issues-- including #21, #29, #37, #58, #73--
complaining that the CMake build system didn't work on Un*x platforms
(although it was not until #73 that this bug came to light.)
This commit removes all vestiges of Un*x support from the CMake build
system and makes it clear that CMake cannot be used to build
libjpeg-turbo on non-WIN32 platforms. It is our position that CMake
will not be supported on non-WIN32 platforms until/unless the autotools
build system is removed, and this will not happen without broad support
from the community (including major O/S vendors.) If you are in favor
of migrating the entire build system to CMake, then please make your
voice heard by commenting on #56.
Even though tjDecompressToYUV2() is mostly just a wrapper for
tjDecompressToYUVPlanes(), tjDecompressToYUV2() still calls
jpeg_read_header(), so it needs to properly set up the libjpeg error
handler prior to making this call. Otherwise, under very esoteric (and
arguably incorrect) use cases, a program could call tjDecompressToYUV2()
without first checking the JPEG header using tjDecompressHeader3(), and
if the header was corrupt, then the libjpeg API would invoke
my_error_exit(). my_error_exit() would in turn call longjmp() on the
previous value of myerr->setjmp_buffer, which was probably set in a
previous TurboJPEG function, such as tjInitDecompress(). Thus, when a
libjpeg error was triggered within the body of tjDecompressToYUV2(), the
PC would jump to the error handler of the previous TurboJPEG function,
and this usually caused stack corruption in the calling program (because
the signature and return type of the previous TurboJPEG function
probably wasn't the same as that of tjDecompressToYUV2().)
Actually, what happened was that the longjmp() call within
my_error_exit() acted on the previous value of myerr->setjmp_buffer,
which was probably set in a previous TurboJPEG function, such as
tjInitDecompress(). Thus, when a libjpeg error was triggered within
the body of tjDecompressToYUV2(), the PC jumped to the error handler
of the previous TurboJPEG function, and this usually caused stack
corruption in the calling program (because the signature and return
type of the previous TurboJPEG function probably wasn't the same.)
Even though tjDecompressToYUV2() is mostly just a wrapper for
tjDecompressToYUVPlanes(), tjDecompressToYUV2() still calls
jpeg_read_header(), so it needs to properly set up the libjpeg error
handler prior to making this call. Otherwise, under very esoteric (and
arguably incorrect) use cases, a program can call tjDecompressToYUV2()
without first checking the JPEG header using tjDecompressHeader3(), and
if the header is corrupt, tjDecompressToYUV2() will abort without
triggering an error.
Fixes#72
<sigh> GitHub doesn't render indented text the same as my local MarkDown
viewer (MacDown), so it's necessary to indent "... OR ..." by 3 spaces
so both will display it on the same indentation level as "Visual C++
2005 or later" and "MinGW".
Indent "... OR ..." to make it clear that the choice is between Visual
C++ and MinGW, not Visual C++ and MinGW + NASM. Move NASM to the top of
the list to make that even more clear. Make it clear that nasm.exe
should be in the PATH.
Addresses concerns raised in #70
This extends the fix in 6709e4a0cf to
include binary PPM/PGM files, thus preventing a malformed binary
PPM/PGM input file from triggering an overrun of the rescale array and
potentially crashing cjpeg.
Note that this issue affected only cjpeg and not the underlying
libjpeg-turbo libraries, and thus it did not represent a security
threat.
Thanks to @hughdavenport for the discovery.
This is a common practice in other infrastructure libraries, such as
OpenSSL and libpng, because it makes it easy to examine an application
binary and determine which version of the library the application was
linked against.
Closes#66
This prevents a malformed motion-JPEG frame (MJPEG frames lack Huffman
tables) from causing the "fast path" of the Huffman decoder to read
uninitialized memory. Essentially, this is doing the same thing for
MJPEG frames as 43d8cf4d45 did for regular
images.
Running 'make -j{jobs}' on a build that was configured with Java
(--with-java) would previously cause an error:
make: *** No rule to make target `TJExample.class', needed by
`turbojpeg.jar'.
It seems that parallel make doesn't understand that the files in
$(JAVA_CLASSES) are all generated from the same invocation of javac, so
it tries to parallelize the building of those files (which of course
doesn't work.) This patch instead makes turbojpeg.jar depend on
classnoinst.stamp. This effectively creates a synchronization fence,
since that file is only created when all of the class files have been
built.
Fixes#62
The x86-64 SIMD accelerations for Huffman encoding used incorrect
stack math to save xmm8-xmm11 on Windows. This caused TJBench to
always report 1 Mpixel/sec for the compression performance, and it
likely would have caused other application issues as well.
The changes relative to 1.4.x are only cosmetic (using const pointers)
and should not affect API/ABI compatibility, but our practice is to
synchronize the API revision with the most recent release that provides
user-visible changes to the API.
This, in combination with the existing jpeg_skip_scanlines() function,
provides the ability to crop the image both horizontally and vertically
while decompressing (certain restrictions apply-- see libjpeg.txt.)
This also cleans up the documentation of the line skipping feature and
removes the "strip decompression" feature from djpeg, since the new
cropping feature is a superset of it.
Refer to #34 for discussion.
Closes#34
Previously, if a custom value of this variable was specified when
running configure, then that value would be lost if configure was
automatically re-run (as a result of changes to configure.ac, for
instance.)
As a bonus, the NASM variable is now also listed when running
'configure --help', so it is obvious how to override the default
NASM command.
The convention used by libjpeg:
type * variable;
is not very common anymore, because it looks too much like
multiplication. Some (particularly C++ programmers) prefer to tuck the
pointer symbol against the type:
type* variable;
to emphasize that a pointer to a type is effectively a new type.
However, this can also be confusing, since defining multiple variables
on the same line would not work properly:
type* variable1, variable2; /* Only variable1 is actually a
pointer. */
This commit reformats the entirety of the libjpeg-turbo code base so
that it uses the same code formatting convention for pointers that the
TurboJPEG API code uses:
type *variable1, *variable2;
This seems to be the most common convention among C programmers, and
it is the convention used by other codec libraries, such as libpng and
libtiff.
Place the authors in the following order:
* libjpeg-turbo authors (2009-) in descending order of the date of their
most recent contribution to the project, then in ascending order of
the date of their first contribution to the project
* Upstream authors in descending order of the date of the first
inclusion of their code (this indicates that their code serves as the
foundation of this code.)
This also adds Siarhei to the author list, since he contributed ARM SIMD
code both as a Nokia employee and more recently as an independent
developer.
We need to garbage collect between iterations of the outside loop in
bufSizeTest() in order to avoid exhausting the heap when running with
Java 6 (which is still used on Linux to test the 32-bit version of
libjpeg-turbo in automated builds.)
Broken by 46ecffa324.
gas-preprocessor.pl and/or the clang assembler apparently don't like
default values in macro arguments, and we need to use a separate const
section for each function (because of our use of adr, also necessitated
by the broken clang assembler.)
... and only if ThunderX is detected. This can be easily expanded later
on to include other CPUs that are known to suffer from slow LD3/ST3, but
it doesn't make sense to disable LD3/ST3 for all non-Android Linux
platforms just because ThunderX is slow.
Full-color compression speedups relative to previous commits:
Cortex-A53 (Nexus 5X), Android, 64-bit: 1.1-13% (avg. 6.0%)
Cortex-A57 (Nexus 5X), Android, 64-bit: 0.0-22% (avg. 6.3%)
Refer to #47 and #50 for discussion
Closes#50
Note that this commit introduces a similar /proc/cpuinfo parser to that
of the ARM32 implementation. It is used to specifically check whether
the code is running on Cavium ThunderX and, if so, disable the ARM64
SIMD Huffman routines (which slow performance by an average of 8% on
that CPU.)
Based on:
a8c282e5e5
Don't include the all: target as a dependency of the tests when
cross-compiling, and ensure that the files generated by the tests are
removed, even if they were created read-only (or if the tests are being
run on a different type of system that doesn't correctly interpret the
file permissions.) This allows one to easily build the code on one
machine and run 'make test' on another.
When cross-compiling, CMakeLists.txt now generates the CTest script
using relative paths, so that CTest can more easily be executed on a
different machine from the build machine. Furthermore, Windows builds
are now tested using md5cmp, just like on Linux, rather than a CMake
script. This prevents issues with differing CMake locations between
the build and test machines.
This also removes some trailing spaces from the md5cmp code and improves
the readability of the test code in CMakeLists.txt.
jinclude.h can't be safely included multiple times, so instead of
including it in the shared (broken-out) headers, it should instead be
included by the source files that include one or more of those headers.
The accelerated Huffman decoder was previously invoked if there were
> 128 bytes in the input buffer. However, it is possible to construct a
JPEG image with Huffman blocks > 430 bytes in length
(http://stackoverflow.com/questions/2734678/jpeg-calculating-max-size).
While such images are pathological and could never be created by a
JPEG compressor, it is conceivable that an attacker could use such an
artifially-constructed image to trigger an input buffer overrun in the
libjpeg-turbo decompressor and thus gain access to some of the data on
the calling program's heap.
This patch simply increases the minimum buffer size for the accelerated
Huffman decoder to 512 bytes, which should (hopefully) accommodate any
possible input.
This addresses a major issue (LJT-01-005) identified in a security audit
by Cure53.
Because of the exposed nature of the libjpeg API, alloc_small() and
alloc_large() can potentially be called by external code. If an
application were to call either of those functions with
sizeofobject > SIZE_MAX - ALIGN_SIZE - 1, then the math in
round_up_pow2() would wrap around to zero, causing that function to
return a small value. That value would likely not exceed
MAX_ALLOC_CHUNK, so the subsequent size checks in alloc_small() and
alloc_large() would not catch the error.
A similar problem could occur in 32-bit builds if alloc_sarray() were
called with
samplesperrow > SIZE_MAX - (2 * ALIGN_SIZE / sizeof(JSAMPLE)) - 1
This patch simply ensures that the size argument to the alloc_*()
functions will never exceed MAX_ALLOC_CHUNK (1 billion). If it did,
then subsequent size checks would eventually catch that error, so we
are instead catching the error before round_up_pow2() is called.
This addresses a minor concern (LJT-01-001) expressed in a security
audit by Cure53.
This addresses a minor concern (LJT-01-002) expressed in a security
audit by Cure53. _tjInitCompress() and _tjInitDecompress() call
(respectively) jpeg_mem_dest_tj() and jpeg_mem_src_tj() with a pointer
to a dummy buffer, in order to set up the destination/source manager.
The dummy buffer should never be used, but it's still better to make it
static so that the pointer in the destination/source manager always
points to a valid region of memory.
Document the latest benchmarks on the Nexus 5X and change the "2-4x"
overall claim to "2-6x". The peak performance on x86 platforms was
already closer to 5x, and the addition of SIMD-accelerated Huffman
encoding gave it that extra push over the cliff.
There aren't really any best practices to follow here. I tried as best
as I could to adopt a standard that would ease any future maintenance
burdens. The basic tenets of that standard are:
* Assembly instructions always start on Column 5, and operands always
start on Column 21, except:
- The instruction and operand can be indented (usually by 2 spaces)
to indicate a separate instruction stream.
- If the instruction is within an enclosing .if block in a macro,
it should always be indented relative to the .if block.
* Comments are placed with an eye toward readability. There are always
at least 2 spaces between the end of a line of code and the associated
in-line comment. Where it made sense, I tried to line up the comments
in blocks, and some were shifted right to avoid overlap with
neighboring instruction lines. Not an exact science.
* Assembler directives and macros use 2-space indenting rules. .if
blocks are indented relative to the macro, and code within the .if
blocks is indented relative to the .if directive.
* No extraneous spaces between operands. Lining up the operands
vertically did not really improve readability-- personally, I think it
made it worse, since my eye would tend to lose its place in the
uniform columns of characters. Also, code with a lot of vertical
alignment is really hard to maintain, since changing one line could
necessitate changing a bunch of other lines to avoid spoiling the
alignment.
* No extraneous spaces in #defines or other directives. In general, the
only extraneous spaces (other than indenting spaces) are between:
- Instructions and operands
- Operands and in-line comments
This standard should be more or less in keeping with other formatting
standards used within the project.
Decompression speedup relative to libjpeg-turbo 1.4.2 (ISLOW IDCT):
48-core ThunderX (RunAbove ARM Cloud), Linux, 64-bit: 60-113% (avg. 86%)
Cortex-A53 (Nexus 5X), Android, 64-bit: 6.8-27% (avg. 14%)
Cortex-A57 (Nexus 5X), Android, 64-bit: 2.0-14% (avg. 6.8%)
Decompression speedup relative to libjpeg-turbo 1.4.2 (IFAST IDCT):
48-core ThunderX (RunAbove ARM Cloud), Linux, 64-bit: 51-98% (avg. 75%)
Minimal speedup (1-5%) observed on iPhone 5S (Cortex-A7)
NOTE: This commit avoids the st3 instruction for non-Android and
non-Apple builds, which may cause a performance regression against
libjpeg-turbo 1.4.x on ARM64 systems that are running plain Linux.
Since ThunderX is the only platform known to suffer from slow ld3 and
st3 instructions, it is probably better to check for the CPU type
at run time and disable ld3/st3 only if ThunderX is detected.
This commit also enables the use of ld3 on Android platforms, which
should be a safe bet, at least for now. This speeds up compression on
the afore-mentioned Nexus Cortex-A53 by 5.5-19% (avg. 12%) and on the
Nexus Cortex-A57 by 1.2-14% (avg. 6.3%), relative to the previous
commits.
This commit also removes unnecessary macros.
Refer to #52 for discussion.
Closes#52.
Based on:
6bad905034488dd7bf174f4d057c1fd3198afc43
* Include information on how to do a 64-bit ARMv8 build with the latest
NDK
* Suggest -fPIE and -pie as default CFLAGS (required for android-16 and
later.
* Remove -fstrict-aliasing flag (-Wall already includes it)
* Include information on how to do a 64-bit ARMv8 build with the latest
NDK
* Suggest -fPIE and -pie as default CFLAGS (required for android-16 and
later.
* Remove -fstrict-aliasing flag (-Wall already includes it)
For whatever reason, the "write" global variable in tjbench.c was
overriding the linkage with the write() system function. This may have
affected other platforms as well but was not known to.
This allows a project to use PKG_CHECK_MODULES() in its configure.ac
file to easily check for the presence of libjpeg-turbo and modify the
compiler/linker flags accordingly. Note that if a project relies solely
on pkg-config to check for libjpeg-turbo, then it will not be possible
to build that project using libjpeg or an earlier version of
libjpeg-turbo.
Closes#53
Based on:
4967138719
Per @ssvb:
ThunderX is an ARM64 chip that dedicates most of its transistor real
estate to providing 48 cores, so each core is not as fast as a result.
Each core is dual-issue & in-order for scalar instructions and has only
a single-issue half-width NEON unit, so the peak throughput is one
128-bit instruction per 2 cycles. So careful instruction scheduling is
important. Furthermore, ThunderX has an extremely slow implementation
of ld2 and ld3, so this commit implements the equivalent of those
instructions using ld1.
Compression speedup relative to libjpeg-turbo 1.4.2:
48-core ThunderX (RunAbove ARM Cloud), Linux, 64-bit: 58-85% (avg. 74%)
relative to jpeg-6b: 1.75-2.14x (avg. 1.95x)
Refer to #49 and #51 for discussion.
Closes#51.
This commit also wordsmiths the ChangeLog entry (the ARMv8 SIMD
implementation is "complete" only for compression-- it still lacks some
decompression algorithms, as does the ARMv7 implementation.)
Based on:
9405b5fd03
which is based on:
f561944ff7962c8ab21f
This adds 64-bit NEON coverage for all of the algorithms that are
covered by the 32-bit NEON implementation, except for h2v1 (4:2:2) fancy
upsampling (used when decompressing 4:2:2 JPEG images.) It also adds
64-bit NEON SIMD coverage for:
* slow integer forward DCT (compressor)
* h2v2 (4:2:0) downsampling (compressor)
* h2v1 (4:2:2) downsampling (compressor)
which are not covered in the 32-bit implementation.
Compression speedups relative to libjpeg-turbo 1.4.2:
Apple A7 (iPhone 5S), iOS, 64-bit: 113-150% (reported)
48-core ThunderX (RunAbove ARM Cloud), Linux, 64-bit: 2.1-33% (avg. 15%)
Refer to #44 and #49 for discussion
This commit also removes the unnecessary
if (simd_support & JSIMD_ARM_NEON)
statements from the jsimd* algorithm functions. Since the jsimd_can*()
functions check for the existence of NEON, the corresponding algorithm
functions will never be called if NEON isn't available.
Based on:
dcd9d84f10b0d87b811f70cd5c8a493e58d9a064837b19542f73dc43ccc8a82b71a261c1b1188c21305c89284e7f443f99954c2b53b77d
Unified version with fixes:
1004a3cd05
Full-color compression speedups relative to libjpeg-turbo 1.4.2:
800 MHz ARM Cortex-A9, iOS, 32-bit: 26-44% (avg. 32%)
Refer to #42 and #47 for discussion.
This commit also removes the unnecessary
if (simd_support & JSIMD_ARM_NEON)
statements from the jsimd* algorithm functions. Since the jsimd_can*()
functions check for the existence of NEON, the corresponding algorithm
functions will never be called if NEON isn't available. Removing those
if statements improved performance across the board by a couple of
percent.
Based on:
fc023c880c
Partially reverts 54014d9c2a. When
building from a git sandbox, as opposed to from an official source
tarball, it is still necessary to run autoreconf.
Closes#48
The Linux build machine has been upgraded to autoconf 2.69, automake
1.15, m4 1.4.17, and libtool 2.4.6, so it is no longer necessary to
recommend running autoreconf prior to building the source, if one is
building from an official source tarball (as opposed to from a git
sandbox.) Also, there is no SVN repository anymore (oops.)
Full-color compression speedups relative to libjpeg-turbo 1.4.2:
2.8 GHz Intel Xeon W3530, Linux, 64-bit: 2.2-18% (avg. 9.5%)
2.8 GHz Intel Xeon W3530, Linux, 32-bit: 10-25% (avg. 17%)
2.3 GHz AMD A10-4600M APU, Linux, 64-bit: 4.9-17% (avg. 11%)
2.3 GHz AMD A10-4600M APU, Linux, 32-bit: 8.8-19% (avg. 15%)
3.0 GHz Intel Core i7, OS X, 64-bit: 3.5-16% (avg. 10%)
3.0 GHz Intel Core i7, OS X, 32-bit: 4.8-14% (avg. 11%)
2.6 GHz AMD Athlon 64 X2 5050e:
Performance-neutral (give or take a few percent)
Full-color compression speedups relative to IPP:
2.8 GHz Intel Xeon W3530, Linux, 64-bit: 4.8-34% (avg. 19%)
2.8 GHz Intel Xeon W3530, Linux, 32-bit: -19%-7.0% (avg. -7.0%)
Refer to #42 for discussion. Numerous other approaches were attempted,
but this one proved to be the most performant across all platforms.
This commit also fixes#3 (works around, really-- the clang-compiled version
of jchuff.c still performs 20% worse than its GCC-compiled counterpart, but
that code is now bypassed by the new SSE2 Huffman algorithm.)
Based on:
2cb4d4133036c94e050d
Traditionally, the x86-64 code did not call init_simd() because it had
no need to (only SSE2 was supported.) However, having the ability to
disable SIMD at run time is a useful testing tool, and all of the other
SIMD implementations have this ability.
Fix a regression introduced in 1.4.1 that prevented 32-bit and 64-bit
libjpeg-turbo RPMs from being installed simultaneously on recent Red
Hat/Fedora distributions. This was due to the addition of the
SIZEOF_SIZE_T macro in jconfig.h, which allows the Huffman codec to
determine the word size at compile time. Since that macro differs
between 32-bit and 64-bit builds, this caused a conflict between the
i386 and x86_64 RPMs (any differing files, other than executables, are
not allowed when 32-bit and 64-bit RPMs are installed simultaneously.)
Since the macro is used only internally, it has been moved into
jconfigint.h.
The unnecessary .arch directive was removed from the ARM64 SIMD code
in d70a5c12fc, thus allowing clang's
integrated assembler to assemble the code on Linux systems. However,
this broke the detection mechanism in acinclude.m4 that tells the build
system whether it needs to use gas-preprocessor.pl. Since one of the
primary motivators for using gas-preprocessor.pl with ARM64 builds is
the lack of .req/.unreq directives in Apple's implementation of clang,
acinclude.m4 now checks whether .req/.unreq can be properly assembled
and uses gas-preprocessor.pl if not.
Closes#33.
Quality values > 95 are not useless. They just may not provide as good
of a size vs. perceptual quality tradeoff as lower quality values. This
also displays the default quality value in the cjpeg usage.
Closes#39
Most of these involved overrunning the signed 32-bit JLONG type whenever
building libjpeg-turbo with a 32-bit compiler. These issues are not
believed to represent actual security threats, but eliminating them
makes it easier to detect such threats should they arise in the future.
These days, INT32 is a commonly-defined datatype in system headers. We
cannot eliminate the definition of that datatype from jmorecfg.h, since
the INT32 typedef has technically been part of the libjpeg API since
version 5 (1994.) However, using INT32 internally is risky, because the
inclusion of a particular header (Xmd.h, for instance) could change the
definition of INT32 from long to int on 64-bit platforms and thus change
the internal behavior of libjpeg-turbo in unexpected ways (for instance,
failing to correctly set __INT32_IS_ACTUALLY_LONG to match the INT32
typedef-- perhaps as a result of including the wrong version of
jpeglib.h-- could cause libjpeg-turbo to produce incorrect results.)
The library has always been built in environments in which INT32 is
effectively long (on Windows, long is always 32-bit, so effectively it's
the same as int), so it makes sense to turn INT32 into an explicitly
long datatype. This ensures that libjpeg-turbo will always behave
consistently, regardless of the headers included at compile time.
Addresses a concern expressed in #26.
The IJG README file has been renamed to README.ijg, in order to avoid
confusion (many people were assuming that that was our project's README
file and weren't reading README-turbo.txt) and to lay the groundwork for
markdown versions of the libjpeg-turbo README and build instructions.
Most of these involved left shifting a negative number, which is
technically undefined (although every modern compiler I'm aware of
will implement this by treating the signed integer as a 2's complement
unsigned integer-- the LEFT_SHIFT() macro just makes this behavior
explicit in order to shut up ubsan.) This also fixes a couple of
non-issues in the entropy codecs, whereby the sanitizer reported an
out-of-bounds index in the 4th argument of jpeg_make_d_derived_tbl().
In those cases, the index was actually out of bounds (caused by a
malformed JPEG image), but jpeg_make_d_derived_tbl() would have caught
the error and aborted prior to actually using the invalid address. Here
again, the fix was to make our intentions explicit so as to shut up
ubsan.
The DSPr2 code was errantly comparing the residual (t9, width & 0xF)
with the end pointer (t4, out + width) instead of the width directly
(a1). This would give the wrong results with any image whose output
width was less than 16. The other small changes (ulw to lw and removal
of the nop) are just some easy optimizations around this code.
This issue caused a buffer overrun and subsequent segfault on images
whose scaled output height was 1 pixel and whose scaled output width was
< 16 pixels. Note that the "plain" (non-fancy and non-merged) upsample
routine, which was affected by this bug, is normally not used except
when decompressing a non-YCbCr JPEG image, but it is also used when
decompressing a single-row image (because the other upsampling
algorithms require at least two rows.)
Closes#16.
(descriptions cribbed by DRC from discussion in #20)
In the x86-64 ABI, the high (unused) DWORD of a 32-bit argument's
register is undefined, so it was incorrect to use a 64-bit mov
instruction to transfer a JDIMENSION argument in the 64-bit SSE2 SIMD
functions. The code worked thus far only because the existing compiler
optimizers weren't smart enough to do anything else with the register in
question, so the upper 32 bits happened to be all zeroes-- for the past
6 years, on every x86-64 compiler previously known to mankind.
The bleeding-edge Clang/LLVM compiler has a smarter optimizer, and
under certain circumstances, it will attempt to load-combine adjacent
32-bit integers from one of the libjpeg structures into a single 64-bit
integer and pass that 64-bit integer as a 32-bit argument to one of the
SIMD functions (which is allowed by the ABI, since the upper 32 bits of
the 32-bit argument's register are undefined.) This caused the
libjpeg-turbo regression tests to crash.
Also enhance the documentation of JDIMENSION to explain that its size
is significant to the implementation of the SIMD code.
Closes#20. Refer also to http://crbug.com/532214.
Previously this information was found in a page on libjpeg-turbo.org,
but there was still some confusion, because README-turbo.txt wasn't
clear as to which license applied to what.
With certain images, compressing using quality=100 and the fast integer
forward DCT will cause the divisor passed to compute_reciprocal() to be
1. In those cases, the library already disables the SIMD quantization
algorithm to avoid 16-bit overflow. However, compute_reciprocal()
doesn't properly handle the divisor==1 case, so we need to use special
values in that case so that the C quantization algorithm will behave
like an identity function.
When compiled with -mfpxx (which is now the default on Debian), there are
some restrictions on the use of odd-numbered FP registers. More details
about FPXX can be found here:
https://dmz-portal.mips.com/wiki/MIPS_O32_ABI_-_FR0_and_FR1_Interlinking
This commit simply changes all uses of FP registers to an even-numbered
equivalent like this:
f0 -> f0
f1 -> f2
f2 -> f4
...
f8 -> f16
This commit should have no observable effect except that the MIPS assembly
will now compile with -mfpxx.
Closes#11
rdbmp.c used the ambiguous INT32 datatype, which is sometimes typedef'ed
to long. Windows bitmap headers use 32-bit signed integers for the
width and height, because height can sometimes be negative (this
indicates a top-down bitmap.) If biWidth or biHeight was negative and
INT32 was a 64-bit long, then biWidth and biHeight were read as a
positive integer > INT32_MAX, which failed the test in line 385:
if (biWidth <= 0 || biHeight <= 0)
ERREXIT(cinfo, JERR_BMP_EMPTY);
This commit refactors rdbmp.c so that it uses the datatypes specified by
Microsoft for the Windows BMP header.
This closes#9 and also provides a better solution for mozilla/mozjpeg#153.
This reassures the caller that the buffers will not be modified and also
allows read-only buffers to be passed to the functions.
Partially reverts 3947a19f25fc8186d3812dbcf8e70baea36ef652.
Declare inbuffer arg in jpeg_mem_src() to be const
This reassures the caller that the buffer will not be modified and also
allows read-only buffers to be passed to the function.
rdbmp.c used the ambiguous INT32 datatype, which is sometimes typedef'ed
to long. Windows bitmap headers use 32-bit signed integers for the
width and height, because height can sometimes be negative (this
indicates a top-down bitmap.) If biWidth or biHeight was negative and
INT32 was a 64-bit long, then biWidth and biHeight were read as a
positive integer > INT32_MAX, which failed the test in line 385:
if (biWidth <= 0 || biHeight <= 0)
ERREXIT(cinfo, JERR_BMP_EMPTY);
This commit refactors rdbmp.c so that it uses the datatypes specified by
Microsoft for the Windows BMP header.
This closes#9 and also provides a better solution for mozilla/mozjpeg#153.
NASM 2.11.08 has a bug that prevents it from properly assembling a
macho64 version of libjpeg-turbo (the resulting binary generates corrupt
images.) 2.11.09 works properly. YASM also works properly and has been
a supported alternative since libjpeg-turbo 1.2.
NASM 2.11.08 has a bug that prevents it from properly assembling a
macho64 version of libjpeg-turbo (the resulting binary generates corrupt
images.) 2.11.09 works properly. YASM also works properly and has been
a supported alternative since libjpeg-turbo 1.2.
NASM 2.11.08 has a bug that prevents it from properly assembling a
macho64 version of libjpeg-turbo (the resulting binary generates corrupt
images.) 2.11.09 works properly. YASM also works properly and has been
a supported alternative since libjpeg-turbo 1.2.
NASM 2.11.08 has a bug that prevents it from properly assembling a
macho64 version of libjpeg-turbo (the resulting binary generates corrupt
images.) 2.11.09 works properly. YASM also works properly and has been
a supported alternative since libjpeg-turbo 1.2.
Under very rare circumstances, decompressing specific corrupt JPEG
images would create a situation whereby GET_BITS(1) was invoked
from within HUFF_DECODE_FAST() when bits_left=0. This produced a right
shift by a negative number of bits, which is undefined in C.
Under very rare circumstances, decompressing specific corrupt JPEG
images would create a situation whereby GET_BITS(1) was invoked
from within HUFF_DECODE_FAST() when bits_left=0. This produced a right
shift by a negative number of bits, which is undefined in C.
Under very rare circumstances, decompressing specific corrupt JPEG
images would create a situation whereby GET_BITS(1) was invoked
from within HUFF_DECODE_FAST() when bits_left=0. This produced a right
shift by a negative number of bits, which is undefined in C.
When using context-based upsampling, use a dummy color conversion
routine instead of a dummy row buffer. This improves performance
(since the actual color conversion routine no longer has to be called),
and it also fixes valgrind errors when decompressing to RGB565.
Valgrind previously complained, because using the RGB565 color
converter with the dummy row buffer was causing a table lookup with
undefined indices.
Under very rare circumstances, decompressing specific corrupt JPEG
images would create a situation whereby GET_BITS(1) was invoked
from within HUFF_DECODE_FAST() when bits_left=0. This produced a right
shift by a negative number of bits, which is undefined in C.
Use a new checked exception type (TJException) when passing through
errors from the underlying C library. This gives the application a
choice of catching all exceptions or just those from TurboJPEG.
Throw IllegalArgumentException at the JNI level when arguments to the
JNI function are incorrect, and when one of the TurboJPEG "utility"
functions returns an error (because, per the C API specification, those
functions will only return an error if one of their arguments is out of
range.)
Remove "throws Exception" from the signature of any methods that no
longer pass through an error from the TurboJPEG C library.
Credit Viktor for the new code
Code formatting tweaks
Change the behavior of the bailif0() macro in the JNI wrapper so that it doesn't throw an exception for an unexpected NULL condition. In fact, in all cases, the underlying JNI API function (such as GetFieldID(), etc.) will throw an Error on its own whenever it returns NULL, so our custom exceptions were never being thrown in that case anyhow. All we need to do is just detect the error and bail out of the C code.
This also corrects a couple of formatting issues (semicolons aren't needed at the end of class definitions, and @Override should be specified for the methods we're overriding from super-classes, so the compiler can sanity-check that we're actually overriding a method and not declaring a new one.)
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1595 632fc199-4ca6-4c93-a231-07263d6284db
-- Use macros to represent the fast FDCT constants, to facilitate comparing the AltiVec implementation of the algorithm with the SSE2 implementation.
-- Rename slow FDCT constants for consistency.
-- Use vec_sra() in all cases in the slow FDCT code. The SSE2 implementation uses psraw, which is an arithmetic shift, so we need to do likewise with AltiVec. Using vec_sr() hasn't caused any problems yet, but it is conceivable that it might cause different behavior in certain corner cases.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1444 632fc199-4ca6-4c93-a231-07263d6284db
This patch also removes an unneeded macro from jdmerge.c.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/branches/1.4.x@1403 632fc199-4ca6-4c93-a231-07263d6284db
This patch also removes an unneeded macro from jdmerge.c.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1402 632fc199-4ca6-4c93-a231-07263d6284db
-----
aee36252be.patch
From aee36252be20054afce371a92406fc66ba6627b5 Mon Sep 17 00:00:00 2001
From: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Date: Wed, 13 Aug 2014 03:50:22 +0300
Subject: [PATCH] ARM: Faster NEON yuv->rgb conversion for Krait and Cortex-A15
The older code was developed and tested only on ARM Cortex-A8 and ARM Cortex-A9.
Tuning it for newer ARM processors can introduce some speed-up (up to 20%).
The performance of the inner loop (conversion of 8 pixels) improves from
~27 cycles down to ~22 cycles on Qualcomm Krait 300, and from ~20 cycles
down to ~18 cycles on ARM Cortex-A15.
The performance remains exactly the same on ARM Cortex-A7 (~58 cycles),
ARM Cortex-A8 (~25 cycles) and ARM Cortex-A9 (~30 cycles) processors.
Also use larger indentation in the source code for separating two independent
instruction streams.
-----
a5efdbf22c.patch
From a5efdbf22ce9c1acd4b14a353cec863c2c57557e Mon Sep 17 00:00:00 2001
From: Siarhei Siamashka <siarhei.siamashka@gmail.com>
Date: Wed, 13 Aug 2014 07:23:09 +0300
Subject: [PATCH] ARM: NEON optimized yuv->rgb565 conversion
The performance of the inner loop (conversion of 8 pixels):
* ARM Cortex-A7: ~55 cycles
* ARM Cortex-A8: ~28 cycles
* ARM Cortex-A9: ~32 cycles
* ARM Cortex-A15: ~20 cycles
* Qualcomm Krait: ~24 cycles
Based on the Linaro rgb565 patch from
https://sourceforge.net/p/libjpeg-turbo/patches/24/
but implements better instructions scheduling.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1385 632fc199-4ca6-4c93-a231-07263d6284db
We can't simply increase JMSG_LENGTH_MAX, because it is part of the libjpeg API, and it is generally assumed that a buffer of this length will be passed to format_message(). Thus, the easiest solution is simply to use a shorter copyright string for JMSG_COPYRIGHT.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/branches/1.2.x@1320 632fc199-4ca6-4c93-a231-07263d6284db
We can't simply increase JMSG_LENGTH_MAX, because it is part of the libjpeg API, and it is generally assumed that a buffer of this length will be passed to format_message(). Thus, the easiest solution is simply to use a shorter copyright string for JMSG_COPYRIGHT.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/branches/1.3.x@1319 632fc199-4ca6-4c93-a231-07263d6284db
We can't simply increase JMSG_LENGTH_MAX, because it is part of the libjpeg API, and it is generally assumed that a buffer of this length will be passed to format_message(). Thus, the easiest solution is simply to use a shorter copyright string for JMSG_COPYRIGHT.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1318 632fc199-4ca6-4c93-a231-07263d6284db
-- Auto-generates HAVE_LOCALE_H macro and adds it to jconfig.h (this is used by rdjpgcom.c.)
-- Reconciles the description and ordering of macros between config.h.in and jconfig.h.in, so the two files can be easily diffed.
-- Eliminates the use of the autoheader-generated config.h in the project and moves relevant internal-only macros into a new file, jconfigint.h. This is to avoid "already defined" warnings in files that were including both config.h (to get the internal autotools package information or the INLINE definition) and jconfig.h.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/branches/1.3.x@1258 632fc199-4ca6-4c93-a231-07263d6284db
-- Auto-generates HAVE_LOCALE_H macro and adds it to jconfig.h (this is used by rdjpgcom.c.)
-- Reconciles the description and ordering of macros between config.h.in and jconfig.h.in, so the two files can be easily diffed.
-- Eliminates the use of the autoheader-generated config.h in the project and moves relevant internal-only macros into a new file, jconfigint.h. This is to avoid "already defined" warnings in files that were including both config.h (to get the internal autotools package information or the INLINE definition) and jconfig.h.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1257 632fc199-4ca6-4c93-a231-07263d6284db
delta = cur0 * 2;
cur0 += delta; /* form error * 3 */
errorptr[0] = (FSERROR) (bpreverr0 + cur0);
cur0 += delta; /* form error * 5 */
bpreverr0 = belowerr0 + cur0;
cur0 += delta; /* form error * 7 */
Each time cur0 is incremented by delta, the compiled code doubles the value of delta (WTF?!) Thus, by the time the end of the block is reached, cur0 is equal to 15 times its former self, not 7 times its former self as it should be. At any rate, it was a lot simpler to just refactor the code so that it uses multiplication.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/branches/1.2.x@1255 632fc199-4ca6-4c93-a231-07263d6284db
delta = cur0 * 2;
cur0 += delta; /* form error * 3 */
errorptr[0] = (FSERROR) (bpreverr0 + cur0);
cur0 += delta; /* form error * 5 */
bpreverr0 = belowerr0 + cur0;
cur0 += delta; /* form error * 7 */
Each time cur0 is incremented by delta, the compiled code doubles the value of delta (WTF?!) Thus, by the time the end of the block is reached, cur0 is equal to 15 times its former self, not 7 times its former self as it should be. At any rate, it was a lot simpler to just refactor the code so that it uses multiplication.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/branches/1.3.x@1253 632fc199-4ca6-4c93-a231-07263d6284db
delta = cur0 * 2;
cur0 += delta; /* form error * 3 */
errorptr[0] = (FSERROR) (bpreverr0 + cur0);
cur0 += delta; /* form error * 5 */
bpreverr0 = belowerr0 + cur0;
cur0 += delta; /* form error * 7 */
Each time cur0 is incremented by delta, the compiled code doubles the value of delta (WTF?!) Thus, by the time the end of the block is reached, cur0 is equal to 15 times its former self, not 7 times its former self as it should be. At any rate, it was a lot simpler to just refactor the code so that it uses multiplication.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1251 632fc199-4ca6-4c93-a231-07263d6284db
-- The Mac and Cygwin packages will now be created with the directory structure defined by the configure variables "prefix", "bindir", "libdir", etc., with the exception that the docs are always installed under /usr/share/doc/{package_name}-{version} on Cygwin and /Library/Documentation/{package_name} on Mac.
-- Fixed a duplicate filename warning when generating RPMs with the default prefix of /opt/libjpeg-turbo.
-- Moved the TurboJPEG libraries out of the system directory on Windows and Mac. It is no longer necessary to put them there, since we are not trying to be backward compatible with TurboJPEG/IPP anymore.
-- Fixed an issue whereby building the "installer" target on Windows would not build the Java JAR file, thus causing an error if the JAR had not been previously built.
-- Building the "install" target on Windows will now install libjpeg-turbo into c:\libjpeg-turbo[-gcc][64] (the same directories used by the installers.) This can be overridden by setting CMAKE_INSTALL_PREFIX.
-- The Java classes on all platforms will now look for the JNI library in the directory under which the build/packaging system installs it.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@946 632fc199-4ca6-4c93-a231-07263d6284db
If the default prefix (/opt/libjpeg-turbo) is used, then we now always install 32-bit libraries in /opt/libjpeg-turbo/lib32 and 64-bit libraries in /opt/libjpeg-turbo/lib64 instead of trying to conform to the Debian or Red Hat conventions. The RPM and DEB packages will now be created with the directory structure defined by the configure variables "prefix", "bindir", "libdir", etc., with the exception that the docs are always installed under /usr/share/doc/{package_name}-{version}.
git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@944 632fc199-4ca6-4c93-a231-07263d6284db
AC_ARG_VAR(JNI_CFLAGS, [C compiler flags needed to include jni.h (default: -I/System/Library/Frameworks/JavaVM.framework/Headers on OS X, '-I/usr/java/include -I/usr/java/include/solaris' on Solaris, and '-I/usr/java/default/include -I/usr/java/default/include/linux' on Linux)])
AC_MSG_CHECKING([whether to build TurboJPEG/OSS Java wrapper])
AC_ARG_WITH([java],
AC_HELP_STRING([--with-java],[Build Java wrapper for the TurboJPEG/OSS library]))
['tjtransform',['tjtransform',['../structtjtransform.html',1,'tjtransform'],['../group___turbo_j_p_e_g.html#gaa29f3189c41be12ec5dee7caec318a31',1,'tjtransform(): turbojpeg.h'],['../group___turbo_j_p_e_g.html#ga9cb8abf4cc91881e04a0329b2270be25',1,'tjTransform(tjhandle handle, const unsigned char *jpegBuf, unsigned long jpegSize, int n, unsigned char **dstBufs, unsigned long *dstSizes, tjtransform *transforms, int flags): turbojpeg.h']]],
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.