[libpng17] Imported from libpng-1.7.0beta68.tar

are derived from previous versions.
Removed much of the long list of previous versions from png.h and
    libpng.3.

ANNOUNCE

@@ -1,4 +1,5 @@
-Libpng 1.6.30beta02 - April 22, 2017
+
+Libpng 1.7.0beta68 - November 12, 2015
 
 This is not intended to be a public release.  It will be replaced
 within a few weeks by a public version or by another test version.
@@ -8,35 +9,960 @@ Files available for download:
 Source files with LF line endings (for Unix/Linux) and with a
 "configure" script
 
-   1.6.30beta02.tar.xz (LZMA-compressed, recommended)
-   1.6.30beta02.tar.gz
+   1.7.0beta68.tar.xz (LZMA-compressed, recommended)
+   1.7.0beta68.tar.gz
 
 Source files with CRLF line endings (for Windows), without the
 "configure" script
 
-   lp1630b02.7z  (LZMA-compressed, recommended)
-   lp1630b02.zip
+   lp170b68.7z  (LZMA-compressed, recommended)
+   lp170b68.zip
 
 Other information:
 
-   1.6.30beta02-README.txt
-   1.6.30beta02-LICENSE.txt
-   libpng-1.6.30beta02-*.asc (armored detached GPG signatures)
+   1.7.0beta68-README.txt
+   1.7.0beta68-LICENSE.txt
+   libpng-1.7.0beta68-*.asc (armored detached GPG signatures)
 
-Changes since the last public release (1.6.29):
+Changes since the last public release (1.6.0):
 
-Version 1.6.30beta01 [April 1, 2017]
-  Added missing "$(CPPFLAGS)" to the compile line for c.pic.o in
-    makefile.linux and makefile.solaris-x86 (Cosmin).
-  Revised documentation of png_get_error_ptr() in the libpng manual.
-  Silence clang -Wcomma and const drop warnings (Viktor Szakats).
-  Update Sourceforge URLs in documentation (https instead of http).
+Version 1.7.0alpha01 [December 15, 2012]
+  Started 1.7.0 branch from libpng-1.6.0beta33.
+  Made 8-bit compose and rgb_to_grayscale accuracy improvements.  These
+    changes cause 16-bit arithmetic to be used for 8-bit data in the gamma
+    corrected compose and grayscale operations.  The arithmetic errors have
+    three sources all of which are fixed in this commit:
+    1) 8-bit linear calculations produce massive errors for lower intensity
+       values.
+    2) The old 16-bit "16 to 8" gamma table code erroneously wrote the lowest
+       output value into a table entry which corresponded to multiple output
+       values (so where the value written should have been the closest to the
+       transformed input value.)
+    3) In a number of cases the code to access the 16-bit table did not round;
+       it did a simple shift, which was wrong and made the side effects of (2)
+       even worse.
+  The new gamma code does not have the 16-to-8 problem at the cost of slightly
+    more calculations and the algorithm used to minimize the number of
+    calculations has been extended to all the 16-bit tables; it has advantages
+    for any significant gamma correction.
+  Rearranged png_struct, remove unused members, change png_set_filter handling
+    png_struct members rearranged - partly to reorder to avoid packing, partly
+    to put frequently accessed members at the start and partly to make
+    the grouping more clear. png_set_filter code has been rewritten and the
+    code shared with png_write_start_row moved to a common function.  Comments
+    in png.h have been made more clear. Minor fixes to
+    contrib/libtests/timepng.c and some of the png_*_tRNS logic, including
+    more error detection in png_set_tRNS.
+  Cleaned up USER_LIMITS feature, removing members from png_struct when not
+    required.
+  Rearranged ARM-NEON optimizations to isolate the machine specific code to
+    the hardware subdirectory, and add comments to pngrutil.c so that
+    implementors of other optimizations will know what to do.
+  Fixed cases of unquoted DESTDIR in Makefile.am.
+  Rebuilt Makefile.in, configure, etc., with autoconf-2.69 and automake-1.12.5.
 
-Version 1.6.30beta02 [April 22, 2017]
-  Document need to check for integer overflow when allocating a pixel
-    buffer for multiple rows in contrib/gregbook, contrib/pngminus,
-    example.c, and in the manual (suggested by Jaeseung Choi).
-  Removed reference to the obsolete PNG_SAFE_LIMITS macro in the documentation.
+Version 1.7.0alpha02 [December 17, 2012]
+  Removed functions that were deprecated in libpng-1.6.0:
+    png_reset_zstream(), png_info_init_3(), and png_data_freer() and its
+    associated flags.
+  Removed some duplicated lines from contrib/tools/scale.c and png.c.
+  Changed some instances of png_warning() to png_app_error().
+  Updated some left over "1.6.0beta32" in code sources.
+  Fixed a "png_structp" prototype (should be png_structrp) in arm_init.c
+  Updated the version-number hack in pngvalid.c
+
+Version 1.7.0alpha03 [December 19, 2012]
+  Cleaned up and enhanced the configure option; libpng now correctly
+    links and tests against zlib with a prefix; tests have been clarified; and
+    irrelevant or obsolete things (as defined by the autotools man page) have
+    been removed.
+  Documented new PNG_ABORT behavior in the manual and commentary in pngerror.c
+  Cleaned up whitespace in the synopsis portion of the manpage "libpng.3"
+  Fixed previous support for Z_PREFIX in configure builds, corrected sCAL APIs;
+    some of these use floating point arithmetic so they need to be disabled if
+    floating point arithmetic is switched off.  This is a quiet API change -
+    previously it appeared that the APIs were supported if fixed point
+    arithmetic was used internally, however they required certain APIs (floor,
+    modf, frexp, atof) that are part of C floating point support. Changed
+    png_fixed and the gamma code specific version of the same to avoid floor(),
+    which may be a library function (not an intrinsic). Removed unused #if 0
+    code.
+  Disassembled the version number in scripts/options.awk (necessary for
+    building on SunOs).
+
+Version 1.7.0alpha04 [December 23, 2012]
+  Added scripts/makefile.msys contributed by Christopher M. Wheeler
+  Made default Zlib compression settings be configurable. This adds #defines to
+    pnglibconf.h to control the defaults.
+  Fixed Windows build issues, enabled ARM compilation. Various warnings issued
+    by earlier versions of GCC fixed for Cygwin and Min/GW (which both use old
+    GCCs.) ARM support is enabled by default in zlib.props (unsupported by
+    Microsoft) and ARM compilation is made possible by deleting the check for
+    x86. The test programs cannot be run because they are not signed.
+
+Version 1.7.0alpha05 [December 24, 2012]
+  Discontinued distributing libpng-1.x.x.tar.bz2.
+  Discontinued distributing libpng-1.7.0-1.6.0-diff.txt and similar.
+
+Version 1.7.0alpha06 [January 1, 2013]
+  Fixed 'make distcheck' on SUN OS - libpng.so was not being removed
+
+Version 1.7.0alpha07 [January 10, 2013]
+  Fixed conceivable but difficult to repro overflow. Also added two test
+    programs to generate and test a PNG which should have the problem.
+
+Version 1.7.0alpha08 [January 17, 2013]
+  Corrected previous attempt at overflow detection in png_set_unknown_chunks()
+  (CVE-2013-7353).  Added overflow detection in png_set_sPLT() and
+  png_set_text_2() (CVE-2013-7354).
+
+Version 1.7.0alpha09 [January 21, 2013]
+  Pulled changes to multi-chunk handling from libpng-1.6.0beta40.
+
+Version 1.7.0alpha10 [February 5, 2013]
+  Make symbol prefixing work with the ARM neon optimizations. Also
+    allow pngpriv.h to be included for preprocessor definitions only, so
+    it can be used in non-C/C++ files.
+  Added png_get_palette_max() function.
+  Changed png_size_t to size_t throughout headers, make sRGB check numbers
+    consistent.
+  Fixed ARM support (Mans Rullgard).
+  Removed stray out-of-order #endif and #ifdef (Mans Rullgard).
+  Ported libpng 1.5 options.awk/dfn file handling to 1.7, fixed one bug.
+  Rebuilt Makefile.in, configure, etc., with autoconf-2.69 and automake-1.13.
+  Minor fixes to assembler changes, verified with a dummy .S file
+  Removed cc -E workround, corrected png_get_palette_max API. Tested on
+    SUN OS cc 5.9, which demonstrates the tokenization problem previously
+    avoided by using /lib/cpp.  Since all .dfn output is now protected
+    in double quotes unless it is to be macro substituted the fix should work
+    everywhere.
+
+Version 1.7.0beta01 [February 15, 2013]
+  Enable parallel tests and rearrange TESTS order to take advantage of
+    the massive speed improvements use a make capable of parallel builds
+    on a multi-CPU machine and pass the right arguments to make (-j10000
+    for GNU make) to get the build to run in parallel.
+  Fixed previous support for Z_PREFIX in configure builds, corrected
+    sCAL APIs; some of these use floating point arithmetic so need to be
+    disabled if floating point arithmetic is switched off.  This is a quiet API
+    change - previously it appeared that the APIs were supported if fixed point
+    arithmetic was used internally, however they required certain APIs (floor,
+    modf, frexp, atof) that are part of C floating point support. Changed
+    png_fixed and the gamma code specific version of the same to avoid floor(),
+    which may be a library function (not an intrinsic). Removed unused #if 0
+    code.
+  Fixed make distcheck to skip over the test result files. They were
+    being included because the whole of the 'tests' directory is included,
+    fixed to just include the actual test files and tests/pngstest.  Also
+    reverted the previous erroneous change to remove the ${srcdir} from the
+    front of contrib, scripts, projects and added an explanation of what is
+    going on to Makefile.am
+  Fixed missing dependency in --prefix builds. The intermediate
+    internal 'prefix.h' file can only be generated correctly after pnglibconf.h,
+    however the dependency was not in Makefile.am.  The symptoms are
+    unpredictable depending on the order make chooses to build pngprefix.h and
+    pnglibconf.h, often the error goes unnoticed because there is a system
+    pnglibconf.h to use instead.
+  Accept "," as a separator in pnglibconf.dfa and allow for
+    continuation lines. This adds to the syntax of "option" and "chunk" lines,
+    allowing the elements to be separated by a "," at the end of an element and
+    interpreting a "," at end of line as a continuation - so the following line
+    is read.  The new syntax is optional and solely provided for readability.
+  Enclosed the new png_get_palette_max in #ifdef PNG_GET_PALETTE_MAX_SUPPORTED
+    block, and revised pnglibconf.h and pnglibconf.h.prebuilt accordingly.
+
+Version 1.7.0beta02 [February 18, 2013]
+  Fixed a race condition in the creation of the build 'scripts' directory
+    while building with a parallel make.
+  Use approved/supported Android method to check for NEON, use Linux/POSIX
+    1003.1 API to check /proc/self/auxv avoiding buffer allocation and other
+    library calls (ported from libpng15).
+  Use parentheses more consistently in "#if defined(MACRO)" tests.
+  Folded long lines.
+
+Version 1.7.0beta03 [February 22, 2013]
+  Reenabled code to allow zero length PLTE chunks for MNG.
+  Fixed ALIGNED_MEMORY support.
+  Allow run-time ARM NEON checking to be disabled. A new configure option:
+    --enable-arm-neon=always will stop the run-time checks. New checks
+    within arm/arm_init.c will cause the code not to be compiled unless
+    __ARM_NEON__ is set. This should make it fail safe (if someone asks
+    for it on then the build will fail if it can't be done.)
+  Updated the INSTALL document.
+
+Version 1.7.0beta04 [February 27, 2013]
+  Revised INSTALL to recommend using CPPFLAGS instead of INCLUDES.
+  Revised scripts/makefile.freebsd to respect ZLIBLIB and ZLIBINC.
+  Revised scripts/dfn.awk to work with the buggy MSYS awk that has trouble
+    with CRLF line endings.
+
+Version 1.7.0beta05 [March 5, 2013]
+  Avoid a possible memory leak in contrib/gregbook/readpng.c
+  Changed user chunk callback API to respect global and per chunk defaults.
+    Previously a return of 0 from a user chunk callback would result in the
+    chunk being saved (if this was safe, even if the chunk was unsafe-to-copy);
+    this change respects the defaults set by the application, so unknown chunks
+    can be discarded by default and known-safe ones preserved.
+  Corrected Android builds and corrected libpng.vers with symbol
+    prefixing. This adds an API to set optimization options externally,
+    providing an alternative and general solution for the non-portable
+    run-time tests used by the ARM Neon code.  It also makes those tests
+    compile and link on Android.  The order of settings vs options in
+    pnglibconf.h is reversed to allow settings to depend on options and
+    options can now set (or override) the defaults for settings.
+
+Version 1.7.0beta06 [March 13, 2013]
+  Corrected simplified API default gamma for color-mapped output, added
+    a flag to change default. In 1.6.0 when the simplified API was used
+    to produce color-mapped output from an input image with no gamma
+    information the gamma assumed for the input could be different from
+    that assumed for non-color-mapped output.  In particular 16-bit depth
+    input files were assumed to be sRGB encoded, whereas in the 'direct'
+    case they were assumed to have linear data.  This was an error.  The
+    fix makes the simplified API treat all input files the same way and
+    adds a new flag to the png_image::flags member to allow the
+    application/user to specify that 16-bit files contain sRGB data
+    rather than the default linear.
+  Fixed bugs in the pngpixel and makepng test programs.
+  Fixed CMakelists.txt to allow building a single variant of the library
+    (Claudio Bley):
+  Introduced a PNG_LIB_TARGETS variable that lists all activated library
+    targets.  It is an error if this variable ends up empty, ie. you have
+    to build at least one library variant.
+  Made the *_COPY targets only depend on library targets actually being build.
+  Use PNG_LIB_TARGETS to unify a code path.
+  Changed the CREATE_SYMLINK macro to expect the full path to a file as the
+    first argument. When symlinking the filename component of that path is
+    determined and used as the link target.
+  Use copy_if_different in the CREATE_SYMLINK macro.
+  Eliminated two warnings from the Intel C compiler. The warnings are
+    technically valid, although a reasonable treatment of division would
+    show it to be incorrect.
+
+Version 1.7.0beta07 [April 14, 2013]
+  Updated documentation of 1.5.x to 1.6.x changes in iCCP chunk handling.
+  Fixed incorrect warning of excess deflate data. End condition - the
+    warning would be produced if the end of the deflate stream wasn't read
+    in the last row.  The warning is harmless.
+  Corrected the test on user transform changes on read. It was in the
+    png_set of the transform function, but that doesn't matter unless the
+    transform function changes the rowbuf size, and that is only valid if
+    transform_info is called.
+  Corrected a misplaced closing bracket in contrib/libtests/pngvalid.c
+    (Flavio Medeiros).
+  Corrected length written to uncompressed iTXt chunks (Samuli Suominen).
+
+Version 1.7.0beta08 [April 18, 2013]
+  Added contrib/tools/fixitxt.c, to repair the erroneous iTXt chunk length
+    written by libpng-1.6.0 and 1.6.1.
+  Disallow storing sRGB information when the sRGB is not supported.
+
+Version 1.7.0beta09 [April 24, 2013]
+  Exposed PNG chunk types in png.h
+  Modified png_uint_32 macros/code to make as few assumptions as possible
+  Revised stack marking in arm/filter_neon.S and configure.ac
+  Ensure that NEON filter stuff is completely disabled when switched 'off'.
+    Previously the ARM NEON specific files were still built if the option
+    was switched 'off' as opposed to being explicitly disabled.
+
+Version 1.7.0beta10 [April 24, 2013]
+  Attempt to fix the PNG_ARM_NEON configuration mess.
+
+Version 1.7.0beta11 [April 26, 2013]
+  Test for 'arm*', not just 'arm' in the host_cpu configure variable.
+
+Version 1.7.0beta12 [April 30, 2013]
+  Added png_app_warning for out-of-range unknown chunk index in
+    png_set_unknown_chunk_location().
+  Expanded manual paragraph about writing private chunks.
+  Avoid dereferencing NULL pointer possibly returned from
+     png_create_write_struct() (Andrew Church).
+
+Version 1.7.0beta13 [May 12, 2013]
+  Revised contrib/pngminim/*/makefile to generate pnglibconf.h with the
+    right zlib header files.
+  Separated CPPFLAGS and CFLAGS in contrib/pngminim/*/makefile
+  Updated contrib/pngminus/pnm2png.c (Paul Stewart):
+    Check for EOF
+    Ignore "#" delimited comments in input file to pnm2png.c.
+    Fixed whitespace handling
+    Added a call to png_set_packing()
+    Initialize dimension values so if sscanf fails at least we have known
+      invalid values.
+  Calculate our own zlib windowBits when decoding rather than trusting the
+    CMF bytes in the PNG datastream.
+  Added an option to force maximum window size for inflating, which was
+    the behavior of libpng15 and earlier.
+  Added png-fix-itxt and png-fix-too-far-back to the built programs and
+    removed warnings from the source code and timepng that are revealed as
+    a result.
+  Detect wrong libpng versions linked to png-fix-too-far-back, which currently
+    only works with libpng versions that can be made to reliably fail when
+    the deflate data contains an out-of-window reference.  This means only
+    1.6 and later.
+  Attempt to detect configuration issues with png-fix-too-far-back, which
+    requires both the correct libpng and the correct zlib to function
+    correctly.
+  Check ZLIB_VERNUM for mismatches, enclose #error in quotes
+  Added information in the documentation about problems with and fixes for
+    the bad CRC and bad iTXt chunk situations.
+
+Version 1.7.0beta14 [June 8, 2013]
+  Removed a redundant test in png_set_IHDR().
+  Added set(CMAKE_CONFIGURATION_TYPES ...) to CMakeLists.txt (Andrew Hundt)
+  Deleted set(CMAKE_BUILD_TYPE) block from CMakeLists.txt
+  Enclose the prototypes for the simplified write API in #ifdef STDIO/#endif
+  Make ARM NEON support work at compile time (not just configure time).
+    This moves the test on __ARM_NEON__ into pngconf.h to avoid issues when
+    using a compiler that compiles for multiple architectures at one time.
+  Removed PNG_FILTER_OPTIMIZATIONS and PNG_ARM_NEON_SUPPORTED from
+    pnglibconf.h, allowing more of the decisions to be made internally
+    (pngpriv.h) during the compile.  Without this, symbol prefixing is broken
+    under certain circumstances on ARM platforms.  Now only the API parts of
+    the optimizations ('check' vs 'api') are exposed in the public header files
+    except that the new setting PNG_ARM_NEON_OPT documents how libpng makes the
+    decision about whether or not to use the optimizations.
+  Protect symbol prefixing against CC/CPPFLAGS/CFLAGS useage.
+    Previous iOS/Xcode fixes for the ARM NEON optimizations moved the test
+    on __ARM_NEON__ from configure time to compile time.  This breaks symbol
+    prefixing because the definition of the special png_init_filter_functions
+    call was hidden at configure time if the relevant compiler arguments are
+    passed in CFLAGS as opposed to CC.  This change attempts to avoid all
+    the confusion that would result by declaring the init function even when
+    it is not used, so that it will always get prefixed.
+
+Version 1.7.0beta15 [June 18, 2013]
+  Revised libpng.3 so that "doclifter" can process it.
+
+Version 1.7.0beta16 [July 5, 2013]
+  Revised example.c to illustrate use of PNG_DEFAULT_sRGB and PNG_GAMMA_MAC_18
+    as parameters for png_set_gamma().  These have been available since
+    libpng-1.5.4.
+  Renamed contrib/tools/png-fix-too-far-back.c to pngfix.c and revised it
+    to check all compressed chunks known to libpng.
+  Updated documentation to show default behavior of benign errors correctly.
+  Added perfect hash code generation for lists of PNG chunks. This is
+    a work in progress; checked in for use in pngfix.c
+  Ported ARM no-read patch from libpng16.
+  Ported pngfix.c patches from libpng16.
+
+Version 1.7.0beta17 [August 21, 2013]
+  Revised manual about changes in iTXt chunk handling made in libpng-1.6.0.
+  Added "/* SAFE */" comments in pngrutil.c and pngrtran.c where warnings
+    may be erroneously issued by code-checking applications.
+  Added information about png_set_options() to the manual.
+  Delay calling png_init_filter_functions() until a row with nonzero filter
+    is found.
+
+Version 1.7.0beta18 [September 16, 2013]
+  Fixed inconsistent conditional compilation of png_chunk_unknown_handling()
+    prototype, definition, and usage.  Made it depend on
+    PNG_HANDLE_AS_UNKNOWN_SUPPORTED everywhere.
+
+Version 1.7.0beta19 [September 30, 2013]
+  Reverted the change to unknown handling #defines; the change breaks 'NOREAD'
+    builds.
+  Fixed default behavior of ARM_NEON_API. If the ARM NEON API option is
+    compiled without the CHECK option it defaulted to on, not off.
+  Catch up with recent libpng16 changes; unknown handling and spelling
+    corrections
+  Avoid up-cast warnings in pngvalid.c. On ARM the alignment requirements of
+    png_modifier are greater than that of png_store and as a consequence
+    compilation of pngvalid.c results in a warning about increased alignment
+    requirements because of the bare cast to (png_modifier*).  The code is
+    safe, because the pointer is known to point to a stack allocated
+    png_modifier, but this change avoids the warning.
+  Fixed some grammatical changes and updated the png chunks list.
+
+Version 1.7.0beta20 [October 13, 2013]
+  Made changes for compatibility with automake 1.14:
+    1) Added the 'compile' program to the list of programs that must be cleaned
+       in autogen.sh
+    2) Added 'subdir-objects' which causes .c files in sub-directories to be
+       compiled such that the corresponding .o files are also in the
+       sub-directory.  This is because automake 1.14 warns that the
+       current behavior of compiling to the top level directory may be removed
+       in the future.
+    3) Updated dependencies on pnglibconf.h to match the new .o locations and
+       added all the files in contrib/libtests and contrib/tools that depend
+       on pnglibconf.h
+    4) Added 'BUILD_SOURCES = pnglibconf.h'; this is the automake recommended
+       way of handling the dependencies of sources that are machine generated;
+       unfortunately it only works if the user does 'make all' or 'make check',
+       so the dependencies (3) are still required.
+  Cleaned up (char*) casts of zlib messages. The latest version of the Intel C
+    compiler complains about casting a string literal as (char*), so copied the
+    treatment of z_const from the library code into pngfix.c
+  Simplified error message code in pngunknown. The simplification has the
+    useful side effect of avoiding a bogus warning generated by the latest
+    version of the Intel C compiler (it objects to
+    condition ? string-literal : string-literal).
+  Make autogen.sh work with automake 1.13 as well as 1.14. Do this by always
+    removing the 1.14 'compile' script but never checking for it.
+
+Version 1.7.0beta21 [November 2, 2013]
+  Added ARMv8 support (James Yu <james.yu at linaro.org>).  Added file
+    arm/filter_neon_intrinsics.c; enable with -mfpu=neon.
+  Revised pngvalid to generate size images with as many filters as it can
+    manage, limited by the number of rows.
+  Cleaned up ARM NEON compilation handling. The tests are now in pngpriv.h
+    and detect the broken GCC compilers.
+  Allow clang derived from older GCC versions to use ARM intrinsics. This
+    causes all clang builds that use -mfpu=neon to use the intrinsics code,
+    not the assembler code.  This has only been tested on iOS 7. It may be
+    necessary to exclude some earlier clang versions but this seems unlikely.
+  Changed NEON implementation selection mechanism. This allows assembler
+    or intrinsics to be turned on at compile time during the build by defining
+    PNG_ARM_NEON_IMPLEMENTATION to the correct value (2 or 1).  This macro
+    is undefined by default and the build type is selected in pngpriv.h.
+
+Version 1.7.0beta22 [November 7, 2013]
+  Fixed #include in filter_neon_intrinsics.c and ctype macros. The ctype char
+    checking macros take an unsigned char argument, not a signed char.
+
+Version 1.7.0beta23 [November 24, 2013]
+  Changed #ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED in pngpread.c to
+    #ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED to be consistent with
+    what is in pngpriv.h.
+  Moved prototype for png_handle_unknown() in pngpriv.h outside of
+    the #ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED/#endif block.
+  Added "-Wall" to CFLAGS in contrib/pngminim/*/makefile
+  Conditionally compile some unused functions reported by -Wall in
+    pngminim.
+  Fixed 'minimal' builds. Various obviously useful minimal configurations
+    don't build because of missing contrib/libtests test programs and
+    overly complex dependencies in scripts/pnglibconf.dfa. This change
+    adds contrib/conftest/*.dfa files that can be used in automatic build
+    scripts to ensure that these configurations continue to build.
+  Enabled WRITE_INVERT and WRITE_PACK in contrib/pngminim/encoder.
+  Fixed pngvalid 'fail' function declaration on the Intel C Compiler.
+    This reverts to the previous 'static' implementation and works round
+    the 'unused static function' warning by using PNG_UNUSED().
+
+Version 1.7.0beta24 [December 15, 2013]
+  Removed or marked PNG_UNUSED some harmless "dead assignments" reported
+    by clang scan-build.
+  Changed tabs to 3 spaces in png_debug macros and changed '"%s"m'
+    to '"%s" m' to improve portability among compilers.
+  Changed png_free_default() to free() in pngtest.c
+  Tidied up pngfix inits and fixed pngtest no-write builds.
+  Bookkeeping: Moved functions around (no changes). Moved transform
+    function definitions before the place where they are called so that
+    they can be made static. Move the intrapixel functions and the
+    grayscale palette builder out of the png?tran.c files. The latter
+    isn't a transform function and is no longer used internally, and the
+    former MNG specific functions are better placed in pngread/pngwrite.c
+  Made transform implementation functions static. This makes the internal
+    functions called by png_do_{read|write}_transformations static. On an
+    x86-64 DLL build (Gentoo Linux) this reduces the size of the text
+    segment of the DLL by 1208 bytes, about 0.6%. It also simplifies
+    maintenance by removing the declarations from pngpriv.h and allowing
+    easier changes to the internal interfaces.
+  Handle zero-length PLTE chunk or NULL palette with png_error()
+    instead of png_chunk_report(), which by default issues a warning
+    rather than an error, leading to later reading from a NULL pointer
+    (png_ptr->palette) in png_do_expand_palette(). This is CVE-2013-6954
+    and VU#650142.  Libpng-1.6.1 through 1.6.7 and libpng-1.7.0beta03
+    through 1.7.0beta23 are vulnerable.  Libpng-1.6.0 and earlier do not
+    have the bug.
+
+Version 1.7.0beta25 [December 26, 2013]
+  Merged files with version 1.6.8 that have not changed since then.
+  Rebuilt configure scripts with automake-1.14.1 and autoconf-2.69
+
+Version 1.7.0beta26 [January 1, 2014]
+  Added libpng 1.5 checks to pngvalid.c
+  Merged with 1.5 and 1.6 changes to create a single pngvalid.c
+  Added dSIG to chunk list in png.h
+  Removed #error macro from contrib/tools/pngfix.c (Thomas Klausner).
+  Merged pngrio.c, pngtrans.c, and pngwio.c with libpng-1.6.9
+  Fixed test programs for interlace options. Made pngvalid.c and
+    pngtest.c work correctly when READ_INTERLACING and/or WRITE_INTERLACING
+    are switched off.
+  Changed pngvalid.c to support libpng 1.5, which does not support
+    the PNG_MAXIMUM_INFLATE_WINDOW option, so #define it out when
+    appropriate in pngvalid.c
+  Allow unversioned links created on install to be disabled in configure.
+    In configure builds 'make install' changes/adds links like png.h
+    and libpng.a to point to the newly installed, versioned, files (e.g.
+    libpng17/png.h and libpng17.a). Three new configure options and some
+    rearrangement of Makefile.am allow creation of these links to be disabled.
+
+Version 1.7.0beta27 [January 10, 2014]
+  Removed potentially misleading warning from png_check_IHDR().
+
+Version 1.7.0beta28 [January 20, 2014]
+  Updated scripts/makefile.* to use CPPFLAGS (Cosmin).
+  Added clang attribute support (Cosmin).
+
+Version 1.7.0beta29 [January 30, 2014]
+  Quiet an uninitialized memory warning from VS2013 in png_get_png().
+
+Version 1.7.0beta30 [February 2, 2014]
+  Fixed a large number of instances where PNGCBAPI was omitted from
+    function definitions.
+  Added pngimage test program for png_read_png and png_write_png. This
+    is a work-in-progress; no tests are run automatically at present and
+    the program by virtue of exhaustively testing all the transforms is
+    very slow.
+
+Version 1.7.0beta31 [February 6, 2014]
+  Make png_read_png() and png_write_png() prototypes in png.h depend
+    upon PNG_READ_SUPPORTED and PNG_WRITE_SUPPORTED.
+  Removed dependence on !PNG_READ_EXPAND_SUPPORTED for calling
+    png_set_packing() in png_read_png().
+  Completed full working pngimage test program and added two test scripts.
+  Fixed combination of ~alpha with shift. On read invert alpha, processing
+    occurred after shift processing, which causes the final values to be
+    outside the range that should be produced by the shift. Reversing the
+    order on read makes the two transforms work together correctly and mirrors
+    the order used on write.
+  Do not read invalid sBIT chunks. Previously libpng only checked sBIT
+    values on write, so a malicious PNG writer could therefore cause
+    the read code to return an invalid sBIT chunk, which might lead to
+    application errors or crashes.  Such chunks are now skipped (with
+    chunk_benign_error).
+  Restored a line, "c = b;", that was inadvertently deleted from the
+    PAETH filtering code from libpng-1.7.0beta24/pngrutil.c. Deleted
+    the other instance of "c = b;" which is the one that triggered a
+    scan-build warning.
+  Support builds with unsupported PNG_TRANSFORM_* values.  All of the
+    PNG_TRANSFORM_* values are always defined in png.h and, because they
+    are used for both read and write in some cases, it is not reliable
+    to #if out ones that are totally unsupported. This change adds error
+    detection in png_read_image() and png_write_image() to do a
+    png_app_error() if the app requests something that cannot be done
+    and it adds corresponding code to pngimage.c to handle such options
+    by not attempting to test them.
+
+Version 1.7.0beta32 [February 26, 2014]
+  Moved redefines of png_error(), png_warning(), png_chunk_error(),
+    and png_chunk_warning() from pngpriv.h to png.h to make them visible
+    to libpng-calling applications.
+  Moved OS dependent code from arm/arm_init.c, to allow the included
+    implementation of the ARM NEON discovery function to be set at
+    build-time and provide sample implementations from the current code in the
+    contrib/arm-neon subdirectory. The __linux__ code has also been changed to
+    compile and link on Android by using /proc/cpuinfo, and the old linux code
+    is in contrib/arm-neon/linux-auxv.c.  The new code avoids POSIX and Linux
+    dependencies apart from opening /proc/cpuinfo and is C90 compliant.
+  Check for info_ptr == NULL early in png_read_end() so we don't need to
+    run all the png_handle_*() and depend on them to return if info_ptr == NULL.
+    This improves the performance of png_read_end(png_ptr, NULL) and makes
+    it more robust against future programming errors.
+  Check for __has_extension before using it in pngconf.h, to
+    support older Clang versions (Jeremy Sequoia).
+  Use a user warning handler in contrib/gregbook/readpng2.c instead of default,
+    so warnings will be put on stderr even if libpng has CONSOLE_IO disabled.
+  Added png_ptr->process_mode = PNG_READ_IDAT_MODE in png_push_read_chunk
+    after recognizing the IDAT chunk, which avoids an infinite loop while
+    reading a datastream whose first IDAT chunk is of zero-length.
+    This fixes CERT VU#684412 and CVE-2014-0333.
+  Don't recognize known sRGB profiles as sRGB if they have been hacked,
+    but don't reject them and don't issue a copyright violation warning.
+  Minor editing of contrib/arm-neon/README and contrib/examples/*.c
+  Moved some documentation from png.h to libpng.3 and libpng-manual.txt
+
+Version 1.7.0beta33 [February 27, 2014]
+  Fixed typos in the manual and in scripts/pnglibconf.dfa (CFLAGS -> CPPFLAGS
+    and PNG_USR_CONFIG -> PNG_USER_CONFIG).
+
+Version 1.7.0beta34 [March 17, 2014]
+  Treat CRC error handling with png_set_crc_action(), instead of with
+    png_set_benign_errors(), which has been the case since libpng-1.6.0beta18.
+  Use "if (value != 0)" instead of "if (value)" consistently.
+  Changed ZlibSrcDir from 1.2.5 to 1.2.8 in projects/vstudio.
+  Moved configuration information from the manual to the INSTALL file.
+
+Version 1.7.0beta35 [August 6, 2014]
+  Removed #if/#else/#endif from inside two pow() calls in pngvalid.c because
+    they were handled improperly by Portland Group's PGI-14.1 - PGI-14.3
+    when using its "__builtin_pow()" function.
+  Silence 'unused parameter' build warnings (Cosmin).
+  $(CP) is now used alongside $(RM_F).  Also, use 'copy' instead
+    of 'cp' where applicable, and applied other minor makefile changes.
+  Don't warn about invalid dimensions exceeding user limits.
+  Allow an easy replacement of the default pre-built configuration
+    header with a custom header, via the make PNGLIBCONF_H_PREBUILT
+    macro.
+  Fixed a typo in pngrutil.c, introduced in libpng-1.5.6, that interferes
+    with "blocky" expansion of sub-8-bit interlaced PNG files (Eric Huss).
+  Removed scripts/makefile.elf. It has not worked since libpng-1.5.0beta14
+    due to elimination of the PNG_FUNCTION_EXPORT and PNG_DATA_EXPORT
+    definitions from pngconf.h.
+  Ensure that CMakeLists.txt makes the target "lib" directory before making
+    symbolic link into it (SourceForge bug report #226 by Rolf Timmermans).
+  Added opinion that the ECCN (Export Control Classification Number) for
+    libpng is EAR99 to the README file.
+
+Version 1.7.0beta36 [September 27, 2014]
+  Updated scripts/makefiles from libpng-1.6.13beta04.
+  Minor updates to the manual from libpng-1.6.14beta04.
+  Fixed incorrect handling of the iTXt compression flag in pngrutil.c
+    (bug report by Shunsaku Hirata).  Bug was introduced in libpng-1.6.0.
+
+Version 1.7.0beta37 [September 28, 2014]
+  Merged scripts/*, pngvalid.c and pngfix.c with libpng-1.6.14beta04.
+
+Version 1.7.0beta38 [October 17, 2014]
+  Added "option READ_iCCP enables READ_COMPRESSED_TEXT" to pnglibconf.dfa
+  Removed unused "text_len" parameter from private function png_write_zTXt().
+  Conditionally compile some code in png_deflate_claim(), when
+    PNG_WARNINGS_SUPPORTED and PNG_ERROR_TEXT_SUPPORTED are disabled.
+  Replaced repeated code in pngpread.c with PNG_PUSH_SAVE_BUFFER_IF_FULL.
+  Added "chunk iTXt enables TEXT" and "chunk zTXt enables TEXT"
+    to pnglibconf.dfa.
+  Removed "option READ_COMPRESSED_TEXT enables READ_TEXT" from pnglibconf.dfa,
+    to make it possible to configure a libpng that supports iCCP but not TEXT.
+  Removed "option WRITE_COMPRESSED_TEXT enables WRITE_TEXT" from pnglibconf.dfa
+  Only mark text chunks as written after successfully writing them.
+
+Version 1.7.0beta39 [November 1, 2014]
+  Ported cosmetic changes from libpng-1.6.15beta02.
+
+Version 1.7.0beta40 [November 6, 2014]
+  Made a one-line revision to configure.ac to support ARM on aarch64
+    (bug report by Marcin Juszkiewicz, fix by John Bowler).
+  Use png_get_libpng_ver(NULL) instead of PNG_LIBPNG_VER_STRING in
+    example.c, pngtest.c, and applications in the contrib directory.
+  Avoid out-of-bounds memory access in png_user_version_check().
+  Simplified and future-proofed png_user_version_check().
+  Fixed GCC unsigned int->float warnings. Various versions of GCC
+    seem to generate warnings when an unsigned value is implicitly
+    converted to double. This is probably a GCC bug but this change
+    avoids the issue by explicitly converting to (int) where safe.
+  Free all allocated memory in pngimage. The file buffer cache was left
+    allocated at the end of the program, harmless but it causes memory
+    leak reports from clang.
+  Fixed array size calculations to avoid warnings. At various points
+    in the code the number of elements in an array is calculated using
+    sizeof.  This generates a compile time constant of type (size_t) which
+    is then typically assigned to an (unsigned int) or (int). Some versions
+    of GCC on 64-bit systems warn about the apparent narrowing, even though
+    the same compiler does apparently generate the correct, in-range,
+    numeric constant.  This adds appropriate, safe, casts to make the
+    warnings go away.
+
+Version 1.7.0beta41 [November 7, 2014]
+  Added #ifdef PNG_16BIT_SUPPORTED/#endif around png_log16bit() and
+    png_do_swap().
+  Changed all "#endif /* PNG_FEATURE_SUPPORTED */" to "#endif /* FEATURE */"
+
+Version 1.7.0beta42 [November 20, 2014]
+  The macros passed in the command line to Borland make were ignored if
+    similarly-named macros were already defined in makefiles. This behavior
+    is different from POSIX make and other make programs.  Surround the
+    macro definitions with ifndef guards (Cosmin).
+  Added "-D_CRT_SECURE_NO_WARNINGS" to CFLAGS in scripts/makefile.vcwin32.
+  Removed the obsolete $ARCH variable from scripts/makefile.darwin.
+  Merged clang no-warning fix from libpng-1.6.13: png_digit was defined
+    but never used in pngerror.c.
+
+Version 1.7.0beta43 [December 18, 2014]
+  Added ".align 2" to arm/filter_neon.S to support old GAS assemblers that
+    don't do alignment correctly.
+  Revised Makefile.am and scripts/*.dfn to work with MinGW/MSYS;
+    renamed scripts/*.dfn to scripts/*.c (Bob Friesenhahn and John Bowler).
+  Quiet a "comparison always true" warning in pngstest.c (John Bowler).
+
+Version 1.7.0beta44 [December 23, 2014]
+  Restored a test on width that was removed from png.c at libpng-1.6.9
+    (Bug report by Alex Eubanks, CVE-2015-0973).
+  Fixed an overflow in png_combine_row() with very wide interlaced images
+    (Bug report and fix by John Bowler, CVE-2014-9495).
+  Corrected the width limit calculation in png_check_IHDR().
+  Removed extraneous handling of PNG_SAFE_LIMITS_SUPPORTED from pngconf.h
+
+Version 1.7.0beta45 [December 28, 2014]
+  Eliminated the PNG_SAFE_LIMITS macro and restored the 1-million-column
+    and 1-million-row default limits in pnglibconf.dfa, that can be reset
+    by the user at build time or run time.  This provides a more robust
+    defense against DOS and as-yet undiscovered overflows.
+  Removed user limits from pngfix. Also pass NULL pointers to
+    png_read_row to skip the unnecessary row de-interlace stuff.
+  Added testing of png_set_packing() to pngvalid.c
+
+Version 1.7.0beta46 [January 2, 2015]
+  Regenerated configure scripts in the tar distributions with libtool-2.4.4
+  Implement previously untested cases of libpng transforms in pngvalid.c
+  Fixed byte order in 2-byte filler, in png_do_read_filler().
+  Allow user to call png_get_IHDR() with NULL arguments (Reuben Hawkins).
+
+Version 1.7.0beta47 [January 29, 2015]
+  Made the check for out-of-range values in png_set_tRNS() detect
+    values that are exactly 2^bit_depth, and work on 16-bit platforms.
+  Added PNG_SKIP_sRGB_CHECK_PROFILE choice for png_set_option().
+  Report all IHDR errors before exiting png_check_IHDR().
+  Merged some parts of libpng-1.6.17beta01 and libpng-1.7.0beta47.
+  Added #ifndef __COVERITY__ where needed in png.c, pngrutil.c and
+    pngset.c to avoid warnings about dead code.
+
+Version 1.7.0beta48 [February 7, 2015]
+  Work around one more Coverity-scan dead-code warning.
+  Do not build png_product2() when it is unused.
+
+Version 1.7.0beta49 [February 11, 2015]
+  Combined sub_row, up_row, avg_row, and paeth_row buffers into a
+    single try_row buffer.
+  Added PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED macro, on by default.
+  Consolidated redundant code in pngwutil.c
+  Deal with integer overflow of sum in pngwutil.c
+
+Version 1.7.0beta50 [February 15, 2015]
+  Combined sub_row, up_row, avg_row, and paeth_row buffers into a
+    single try_row buffer and in cases where two or more of those are
+    being tested, a second tst_row buffer.  This improves CPU speed
+    over that achieved by libpng-1.7.0beta49.
+
+Version 1.7.0beta51 [February 17, 2015]
+  Combined similar parts of png_write_find_filter() into a png_increase_lmins()
+    function.
+  Display user limits in the output from pngtest.
+  Replaced test for harmless integer overflow (wraparound) in pngwutil.c
+    with a method that prevents overflow and does not increase cpu usage
+    significantly.
+
+Version 1.7.0beta52 [February 21, 2015]
+  Added information about setjmp/simplified API interaction in the
+    manual page and INSTALL file. Added information about using
+    row_pointers with a single height*width*bpp allocated block
+  Rebuilt configure scripts with automake-1.15 and libtool-2.4.6
+
+Version 1.7.0beta53 [February 23, 2015]
+  Restored compiling of png_reciprocal2 with PNG_NO_16BIT.
+
+Version 1.7.0beta54 [March 4, 2015]
+  Moved png_set_filter() prototype into a PNG_WRITE_SUPPORTED block
+    of png.h.
+  Conditionally compile a small amount of code in pngwutil.c
+  Avoid runtime checks when converting integer to png_byte with
+    Visual Studio (Sergey Kosarevsky)
+  Merged pngwrite.c with libpng-1.6.17beta06/pngwrite.c
+
+Version 1.7.0beta55 [March 9, 2015]
+  Removed some comments that the configure script did not handle
+    properly from scripts/pnglibconf.dfa and pnglibconf.h.prebuilt.
+  Spelling fixes ("then", "occured") reported by Lauri Nurmi.
+  Added more error reporting in png_default_read_data().
+  Merged png_free_data() with libpng-1.6.17rc02.
+
+Version 1.7.0beta56 [March 11, 2015]
+  Updated CMakeLists.txt to add OSX framework, change YES/NO to ON/OFF
+    for consistency, and remove some useless tests (Alexey Petruchik).
+
+Version 1.7.0beta57 [March 16, 2015]
+  Remove pnglibconf.h, pnglibconf.c, and pnglibconf.out instead of
+    pnglibconf.* in "make clean" (Cosmin).
+  Fix bug in calculation of maxbits, in png_write_sBIT, introduced
+    in libpng-1.6.17beta01 (John Bowler).
+  Revert change to png_default_read_data() made in libpng-1.7.0beta55.
+
+Version 1.7.0beta58 [March 25, 2015]
+  Implemented affirm() support and usage.
+  Remove pnglibconf.dfn and pnglibconf.pre with "make clean".
+  Added byte, short and other overflow checking
+  Added 'affirm' based overflow checking for debug builds for char
+    and short assignments. Simply ignore the error in release builds
+    (a truncated value will be used without warning). Controlled by
+    PNG_RANGE_CHECK_SUPPORTED.
+  Improved some overflow handling. PNG_RELEASE_BUILD replaces tests
+    where the code depended on the build base type and can be defined on
+    the command line, allowing testing in beta builds.  Overflows previously
+    handled by png_warning in png.c have been changed to use new
+    macros/affirm functions so that beta builds will abort on overflow and
+    release builds will quietly ignore it.  This avoids release builds
+    producing warnings that are of no use to end users.
+  Check (png_uint_16) casts for overflow. Also removed png_muldiv_warn,
+    added RELEASE/!RELEASE convenience macros. png_muldiv_warn was used in
+    only one place, and the overflow condition is a genuine warning not
+    an internal error.  Four macros allow code or function parameters to be
+    condition on RELEASE (or not) builds and tidy up the #ifdef handling of
+    functions.
+
+Version 1.7.0beta59 [April 19, 2015]
+  Transformed rewrite: changed row_info, added checks. This introduces an
+    internal struct (png_transform_control) to replace row_info and uses
+    that to implement affirms correctly.  The change also adds checks on
+    the rowbytes calculation and additional checks on most transform
+    implementations.
+  Added png_uint_16 range checking, pngvalid tRNS, fixed png_uint_16:
+    review of previous checks, removal of some where SAFE. pngvalid: add
+    testing of tRNS for better code coverage pngvalid: correct rgb-to-gray
+    error calculations. Code coverage is still incomplete: see /*UNTESTED*/
+    in pngrtran.c. This fixes some arithmetic errors that caused some tests
+    to fail on some 32-bit platforms (Bug reports by Peter Breitenlohner [i686]
+    and Petr Gajdos [i586]).
+  Suppressed some warnings from the Borland C++ 5.5.1/5.82 compiler
+    (Bug report by Viktor Szaka'ts).
+
+Version 1.7.0beta60 [May 6, 2015]
+  Replaced "unexpected" with an integer (0xabadca11) in pngset.c
+    where a long was expected, to avoid a compiler warning when PNG_DEBUG > 1.
+
+Version 1.7.0beta61 [May 31, 2015]
+  Avoid Coverity issue 80858 (REVERSE NULL) in pngtest.c
+  Avoid a harmless potential integer overflow in png_XYZ_from_xy() (Bug
+    report from Christopher Ferris).
+  Changed png_voidcast(), etc., to voidcast(), etc., in contrib/tools/pngfix.c
+    to avoid confusion with the libpng private macros.
+  Fixed old cut&paste bug in the weighted filter selection code in
+    pngwutil.c, introduced in libpng-0.95, March 1997.
+
+Version 1.7.0beta62 [June 1, 2015]
+  Removed the experimental WRITE_WEIGHTED_FILTERED code, to save a few
+    kbytes of the compiled library size. It never worked properly and as
+    far as we can tell, no one uses it. The png_set_filter_heuristics() and
+    png_set_filter_heuristics_fixed() APIs are retained but deprecated.
+
+Version 1.7.0beta63 [June 6, 2015]
+  Quieted some Coverity issues in pngfix.c, png-fix-itxt.c, pngvalid.c,
+    pngstest.c, and pngimage.c. Most seem harmless, but png-fix-itxt
+    would only work with iTXt chunks with length 255 or less.
+  Fixed cexcept.h in which GCC 5 reported that one of the auto
+    variables in the Try macro needs to be volatile to prevent value
+    being lost over the setjmp, and fixed g++ build breaks (John Bowler).
+  Moved READ_GAMMA to pngrtran.c. This makes everything in pngrtrans.c
+    depend on READ_TRANSFORMS and moves all the transform gamma (READ_GAMMA)
+    code from elsewhere to png_tran.c.  There are no code changes. 
+    There is one remaining use of the gamma (16-bit) code in the simplified
+    API in pngread.c but that is because of a long-standing libpng bug,
+    namely that the gamma corrected palette is no produced by
+    png_read_update_info (John Bowler).
+  Completed AArch64 support. The previous change to recognize __ARM_NEON
+    didn't touch all the required preprocessor lines. This change removes
+    support for __ARM_NEON__ completely and makes this consistent.
+  Added #ifdef's to contrib/examples programs so people don't try
+    to compile them without the minimum required support enabled
+    (suggested by Flavio Medeiros).
+
+Version 1.7.0beta64 [July 26, 2015]
+  Removed non-working progressive reader 'skip' function. This
+    function has apparently never been used. It was implemented
+    to support back-door modification of png_struct in libpng-1.4.x
+    but (because it does nothing and cannot do anything) was apparently
+    never tested (John Bowler).
+  Eliminated the final two Coverity defects (insecure temporary file
+    handling in contrib/libtests/pngstest.c; possible overflow of
+    unsigned char in contrib/tools/png-fix-itxt.c). To use the "secure"
+    file handling, define PNG_USE_MKSTEMP, otherwise "tmpfile()" will
+    be used.
+  Removed some unused WEIGHTED_FILTER macros from png.h and pngstruct.h
+  Rearranged png.h to put the major sections in the same order as
+    in libpng16.
+  Belatedly added Mans Rullgard and James Yu to the list of Contributing
+    Authors.
+
+Version 1.7.0beta65 [September 16, 2015]
+  Use nanosleep() instead of usleep() in contrib/gregbook/rpng2-x.c
+    because usleep() is deprecated (port from libpng16).
+  Fixed potential leak of png_pixels in contrib/pngminus/pnm2png.c
+  Fixed uninitialized variable in contrib/gregbook/rpng2-x.c
+  Moved config.h.in~ from the "libpng_autotools_files" list to the
+    "libpng_autotools_extra" list in autogen.sh because it was causing a
+    false positive for missing files (bug report by Robert C. Seacord).
+  Fixed some bad links in the man page.
+  Renamed the arm subdirectory to contrib/neon.
+  Made minor fixes to the test scripts:
+    pngimage: add --list-combos pngunknown: add --strict to catch warnings
+    pngvalid-standard: remove the spurious --progressive read (compare with
+    pngvalid-progressive-standard!) (John Bowler).
+  Implemented a new chunk parse implementation that can be shared. It
+    is currently shared by the progressive reader and the sequential one
+    (but not yet by the writer) (John Bowler).
+  Implemented shared transform handling that is used throughout (John Bowler).
+
+Version 1.7.0beta66 [October 2, 2015]
+  Enabled the low-bit-depth gray tests that were disabled in prior versions
+    of libpng because of problems that should have been fixed by the recent
+    changes to libpng17. Enabling the tests revealed bugs in those changes
+    which are now corrected.
+  Replaced the memcpy of png_transform_background with a structure copy
+    which is safer since it doesn't depend on knowing the first member to
+    be copied, also the copies are improved to copy the transform args too;
+    not required at present but it may prevent a bug being introduced in
+    the future.
+  Fixed some new Coverity defects that were introduced in 1.7.0beta65.
+  Fixed makepng.c so that it compiles with GCC 5.1.0.
+  Added contrib/testspngs/: a directory for additional test png files
+    generated by makepng.
+  Fixed dead code reported by Coverity (John Bowler).
+  Limit the calls to crc32 to ZLIB_IO_MAX, although it is not clear that
+    this is optimal (John Bowler).
+  Make pngstest error limits version specific. Splitting the machine
+    generated error structs out to a file allows the values to be updated
+    without changing pngstest.c itself. Since libpng 1.6 and 1.7 have
+    slightly different error limits this simplifies maintenance. The
+    makepngs.sh script has also been updated to more accurately reflect
+    current problems in libpng 1.7 (John Bowler).
+  Fixed gamma-handling issues. One serious bug is that 8 was forced
+    as a bit depth when caching a palette, even though the palette had been
+    expanded.  One less serious bug, the optimization for avoiding gamma
+    correction introduced significant errors if the app subsequently
+    did further corrections, only shown by using files with gAMA 1/1.52 in
+    pngstest (John Bowler).
+  Suppressed bogus Coverity issues in pngrutil.c and pngwutil.c.
+  Corrected sRGB gamma handling in the Simplified API (John Bowler).
+    If a gamma encoded file that has a gamma not matching that of sRGB is
+    passed to the simplified API the previous code simply interpreted it as
+    a power law encoding.  However old Mac files had a power law correction
+    of 1.45 built in to the encoding, even though the display devices were
+    consistent with sRGB.  Assuming a power law encoding results in
+    substantial differences in the interpretation of low 8-bit values;
+    below 10.  For example an Apple '5' which is equivalent to an sRGB '17'
+    ends up as the value '8'.  This patch provides some measure of correction
+    for this by making the gamma correction done within the simplified API
+    assume that any encoded data is encoded relative to an sRGB-like transfer
+    function; the data is corrected back to the PNG-nominal 2.2 value then
+    decoded to linear (if required) using the sRGB transfer function.
+    This reduces the errors reported by pngstest for such files (colormapped
+    ones) but still leaves the issue with files where the standard libpng
+    code does the gamma decoding.  To cope with the latter cases there is a
+    new pngstest-errors which allows the result; however, this is still a
+    work-in-progress; a better solution is possible.
+  Added a '--small' option to 'makepng' to allow the generation of smaller
+    images that also give better test case coverage.  Also added copyright
+    and licence to the generated images, reflecting the fact that they are
+    original works of the author of the code.  The licence used is the
+    Creative Commons Public Domain (CC0-1.0) one, which is appropriate
+    for non-source-code works.  The resultant images were used to generate
+    the recently-updated pngstest-errors.h file (John Bowler).
+  Added test files generated by "contrib/testpngs/makepngs.sh --small"
+    with MAKEPNG set to the makepng compiled from the current
+    contrib/libtests/makepngs.sh, to the "contrib/testpngs" directory.
+  Made png_rgb_to_gray() optimization fixes.  This fixes the code
+    that optimizes RGB to Gray transformations that only selected on
+    channel and clarifies the handling of max_depth in the transform
+    code (John Bowler).
+  Incorporated new test PNG files into make check.  tests/pngstest-*
+    are changed so that the new test files are divided into 8 groups by
+    gamma and alpha channel.  These tests have considerably better code
+    and pixel-value coverage than contrib/pngsuite; however,coverage is
+    still incomplete (John Bowler).
+  Fixed png_set_tRNS_to_alpha for low-bit-depth gray and enabled
+    low-bit-depth gray tests in pngvalid (John Bowler).
+  Fixed sBIT handling in the low-bit-depth gray and pngstest cases,
+    extended sBIT handling to the colormap code in the simplified API
+    which is separately implemented in pngread.c (John Bowler).
+
+Version 1.7.0beta67 [November 3, 2015]
+  Fixed 'pow' macros in pngvalid.c. It is legal for 'pow' to be a
+    macro, therefore the argument list cannot contain preprocessing
+    directives.  Make sure pow is a function where this happens. This is
+    a minimal safe fix, the issue only arises in non-performance-critical
+    code (bug report by Curtis Leach, fix by John Bowler).
+  Added sPLT chunk support to pngtest.c
+  Prevent setting or writing over-length PLTE chunk (Cosmin Truta).
+  Silently truncate over-length PLTE chunk while reading.
+  Fixed some inconsequential cut-and-paste typos in png_set_cHRM_XYZ_fixed().
+  Clarified COPYRIGHT information to state explicitly that versions
+    are derived from previous versions.
+  Removed much of the long list of previous versions from png.h and
+    libpng.3.
+
+Version 1.7.0beta68 [November 12, 2015]
+  Fixed new bug with CRC error after reading an over-length palette
+    (bug report by Cosmin Truta).
+  Cleaned up coding style in png_handle_PLTE().
 
 Send comments/corrections/commendations to png-mng-implement at lists.sf.net
 (subscription required; visit

CMakeLists.txt

@@ -1,33 +1,13 @@
 # CMakeLists.txt
 
-# Copyright (C) 2007,2009-2016 Glenn Randers-Pehrson
-# Written by Christian Ehrlicher, 2007
-# Revised by Roger Lowman, 2009-2010
-# Revised by Clifford Yapp, 2011-2012
-# Revised by Roger Leigh, 2016
-# Revised by Andreas Franek, 2016
+# Copyright (C) 2007-2015 Glenn Randers-Pehrson
 
 # This code is released under the libpng license.
 # For conditions of distribution and use, see the disclaimer
 # and license in png.h
 
-cmake_minimum_required(VERSION 2.8.3)
-cmake_policy(VERSION 2.8.3)
-
-# Set MacOSX @rpath usage globally.
-if (POLICY CMP0020)
-  cmake_policy(SET CMP0020 NEW)
-endif(POLICY CMP0020)
-if (POLICY CMP0042)
-  cmake_policy(SET CMP0042 NEW)
-endif(POLICY CMP0042)
-# Use new variable expansion policy.
-if (POLICY CMP0053)
-  cmake_policy(SET CMP0053 NEW)
-endif(POLICY CMP0053)
-if (POLICY CMP0054)
-  cmake_policy(SET CMP0054 NEW)
-endif(POLICY CMP0054)
+cmake_minimum_required(VERSION 2.4.4)
+set(CMAKE_ALLOW_LOOSE_LOOP_CONSTRUCTS true)
 
 set(CMAKE_CONFIGURATION_TYPES "Release;Debug;MinSizeRel;RelWithDebInfo")
 
@@ -35,23 +15,14 @@ project(libpng C)
 enable_testing()
 
 set(PNGLIB_MAJOR 1)
-set(PNGLIB_MINOR 6)
-set(PNGLIB_RELEASE 30)
+set(PNGLIB_MINOR 7)
+set(PNGLIB_RELEASE 0)
 set(PNGLIB_NAME libpng${PNGLIB_MAJOR}${PNGLIB_MINOR})
 set(PNGLIB_VERSION ${PNGLIB_MAJOR}.${PNGLIB_MINOR}.${PNGLIB_RELEASE})
 
-include(GNUInstallDirs)
-
 # needed packages
-
-#Allow users to specify location of Zlib,
-# Useful if zlib is being built alongside this as a sub-project
-option(PNG_BUILD_ZLIB "Custom zlib Location, else find_package is used" OFF)
-
-IF(NOT PNG_BUILD_ZLIB)
-  find_package(ZLIB REQUIRED)
-  include_directories(${ZLIB_INCLUDE_DIR})
-ENDIF(NOT PNG_BUILD_ZLIB)
+find_package(ZLIB REQUIRED)
+include_directories(${ZLIB_INCLUDE_DIR})
 
 if(NOT WIN32)
   find_library(M_LIBRARY
@@ -76,346 +47,32 @@ option(PNG_FRAMEWORK "Build OS X framework" OFF)
 option(PNG_DEBUG     "Build with debug output" OFF)
 option(PNGARG        "Disable ANSI-C prototypes" OFF)
 
-set(PNG_PREFIX "" CACHE STRING "Prefix to add to the API function names")
-set(DFA_XTRA "" CACHE FILEPATH "File containing extra configuration settings")
-
-# set definitions and sources for arm
-if(${CMAKE_SYSTEM_PROCESSOR} MATCHES "^arm" OR
-  ${CMAKE_SYSTEM_PROCESSOR} MATCHES "^aarch64")
-  set(PNG_ARM_NEON_POSSIBLE_VALUES check on off)
-  set(PNG_ARM_NEON "check" CACHE STRING "Enable ARM NEON optimizations:
-     check: (default) use internal checking code;
-     off: disable the optimizations;
-     on: turn on unconditionally.")
-  set_property(CACHE PNG_ARM_NEON PROPERTY STRINGS
-     ${PNG_ARM_NEON_POSSIBLE_VALUES})
-  list(FIND PNG_ARM_NEON_POSSIBLE_VALUES ${PNG_ARM_NEON} index)
-  if(index EQUAL -1)
-    message(FATAL_ERROR
-      " PNG_ARM_NEON must be one of [${PNG_ARM_NEON_POSSIBLE_VALUES}]")
-  elseif(NOT ${PNG_ARM_NEON} STREQUAL "no")
-    set(libpng_arm_sources
-      arm/arm_init.c
-      arm/filter_neon.S
-      arm/filter_neon_intrinsics.c)
-
-    if(${PNG_ARM_NEON} STREQUAL "on")
-      add_definitions(-DPNG_ARM_NEON_OPT=2)
-    elseif(${PNG_ARM_NEON} STREQUAL "check")
-      add_definitions(-DPNG_ARM_NEON_CHECK_SUPPORTED)
-    endif()
-  else()
-    add_definitions(-DPNG_ARM_NEON_OPT=0)
-  endif()
-endif()
-
-# set definitions and sources for powerpc
-if(${CMAKE_SYSTEM_PROCESSOR} MATCHES "^powerpc*" OR
-	${CMAKE_SYSTEM_PROCESSOR} MATCHES "^ppc64*" )
-  set(PNG_POWERPC_VSX_POSSIBLE_VALUES check on off)
-  set(PNG_POWERPC_VSX "check" CACHE STRING "Enable POWERPC VSX optimizations:
-     check: (default) use internal checking code;
-     off: disable the optimizations;
-     on: turn on unconditionally.")
-  set_property(CACHE PNG_POWERPC_VSX PROPERTY STRINGS
-     ${PNG_POWERPC_VSX_POSSIBLE_VALUES})
-  list(FIND PNG_POWERPC_VSX_POSSIBLE_VALUES ${PNG_POWERPC_VSX} index)
-  if(index EQUAL -1)
-    message(FATAL_ERROR
-      " PNG_POWERPC_VSX must be one of [${PNG_POWERPC_VSX_POSSIBLE_VALUES}]")
-  elseif(NOT ${PNG_POWERPC_VSX} STREQUAL "no")
-    set(libpng_powerpc_sources
-      powerpc/powerpc_init.c
-      powerpc/filter_vsx_intrinsics.c)
-
-    if(${PNG_POWERPC_VSX} STREQUAL "on")
-      add_definitions(-DPNG_POWERPC_VSX_OPT=2)
-    elseif(${PNG_POWERPC_VSX} STREQUAL "check")
-      add_definitions(-DPNG_POWERPC_VSX_CHECK_SUPPORTED)
-      message(WARNING
-      "[PNG_POWERPC_VSX==check] Please check contrib/powerpc/README file for the list of supported OSes.")
-    endif()
-  else()
-    add_definitions(-DPNG_POWERPC_VSX_OPT=0)
-  endif()
-endif()
-
-
 # SET LIBNAME
 set(PNG_LIB_NAME png${PNGLIB_MAJOR}${PNGLIB_MINOR})
 
 # to distinguish between debug and release lib
 set(CMAKE_DEBUG_POSTFIX "d")
 
-include(CheckCSourceCompiles)
-option(ld-version-script "Enable linker version script" ON)
-if(ld-version-script AND NOT APPLE)
-  # Check if LD supports linker scripts.
-  file(WRITE "${CMAKE_CURRENT_BINARY_DIR}/conftest.map" "VERS_1 {
-        global: sym;
-        local: *;
-};
-
-VERS_2 {
-        global: sym2;
-                main;
-} VERS_1;
-")
-  set(CMAKE_REQUIRED_FLAGS_SAVE ${CMAKE_REQUIRED_FLAGS})
-  set(CMAKE_REQUIRED_FLAGS ${CMAKE_REQUIRED_FLAGS} "-Wl,--version-script='${CMAKE_CURRENT_BINARY_DIR}/conftest.map'")
-  check_c_source_compiles("void sym(void) {}
-void sym2(void) {}
-int main(void) {return 0;}
-" HAVE_LD_VERSION_SCRIPT)
-  if(NOT HAVE_LD_VERSION_SCRIPT)
-    set(CMAKE_REQUIRED_FLAGS ${CMAKE_REQUIRED_FLAGS_SAVE} "-Wl,-M -Wl,${CMAKE_CURRENT_BINARY_DIR}/conftest.map")
-  check_c_source_compiles("void sym(void) {}
-void sym2(void) {}
-int main(void) {return 0;}
-" HAVE_SOLARIS_LD_VERSION_SCRIPT)
-  endif()
-  set(CMAKE_REQUIRED_FLAGS ${CMAKE_REQUIRED_FLAGS_SAVE})
-  file(REMOVE "${CMAKE_CURRENT_BINARY_DIR}/conftest.map")
-endif()
-
-# Find symbol prefix.  Likely obsolete and unnecessary with recent
-# toolchains (it's not done in many other projects).
-function(symbol_prefix)
-  set(SYMBOL_PREFIX)
-
-  execute_process(COMMAND "${CMAKE_C_COMPILER}" "-E" "-"
-                  INPUT_FILE /dev/null
-                  OUTPUT_VARIABLE OUT
-                  RESULT_VARIABLE STATUS)
-
-  if(CPP_FAIL)
-    message(WARNING "Failed to run the C preprocessor")
-  endif()
-
-  string(REPLACE "\n" ";" OUT "${OUT}")
-  foreach(line ${OUT})
-    string(REGEX MATCH "^PREFIX=" found_match "${line}")
-    if(found_match)
-      STRING(REGEX REPLACE "^PREFIX=(.*\)" "\\1" prefix "${line}")
-      string(REGEX MATCH "__USER_LABEL_PREFIX__" found_match "${prefix}")
-      if(found_match)
-        STRING(REGEX REPLACE "(.*)__USER_LABEL_PREFIX__(.*)" "\\1\\2" prefix "${prefix}")
-      endif()
-      set(SYMBOL_PREFIX "${prefix}")
-    endif()
-  endforeach()
-
-    message(STATUS "Symbol prefix: ${SYMBOL_PREFIX}")
-    set(SYMBOL_PREFIX "${SYMBOL_PREFIX}" PARENT_SCOPE)
-endfunction()
-
-if(UNIX)
-  symbol_prefix()
-endif()
-
-find_program(AWK NAMES gawk awk)
-
+# Use the prebuilt pnglibconf.h file from the scripts folder
+# TODO: fix this by building with awk; without this no cmake build can be
+# configured directly (to do so indirectly use your local awk to build a
+# pnglibconf.h in the build directory.)
+configure_file(${CMAKE_CURRENT_SOURCE_DIR}/scripts/pnglibconf.h.prebuilt
+               ${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.h)
 include_directories(${CMAKE_CURRENT_BINARY_DIR})
 
-if(NOT AWK)
-  # No awk available to generate sources; use pre-built pnglibconf.h
-  configure_file(${CMAKE_CURRENT_SOURCE_DIR}/scripts/pnglibconf.h.prebuilt
-                 ${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.h)
-  add_custom_target(genfiles) # Dummy
-else()
-  include(CMakeParseArguments)
-  # Generate .chk from .out with awk
-  # generate_chk(INPUT inputfile OUTPUT outputfile [DEPENDS dep1 [dep2...]])
-  function(generate_chk)
-    set(options)
-    set(oneValueArgs INPUT OUTPUT)
-    set(multiValueArgs DEPENDS)
-    cmake_parse_arguments(_GC "${options}" "${oneValueArgs}" "${multiValueArgs}" ${ARGN})
-    if (NOT _GC_INPUT)
-      message(FATAL_ERROR "Invalid arguments.  generate_out requires input.")
-    endif()
-    if (NOT _GC_OUTPUT)
-      message(FATAL_ERROR "Invalid arguments.  generate_out requires output.")
-    endif()
-
-    add_custom_command(OUTPUT "${_GC_OUTPUT}"
-                       COMMAND "${CMAKE_COMMAND}"
-                               "-DINPUT=${_GC_INPUT}"
-                               "-DOUTPUT=${_GC_OUTPUT}"
-                               -P "${CMAKE_CURRENT_BINARY_DIR}/scripts/genchk.cmake"
-                       DEPENDS "${_GC_INPUT}" ${_GC_DEPENDS}
-                       WORKING_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}")
-  endfunction()
-
-  # Generate .out from .c with awk
-  # generate_out(INPUT inputfile OUTPUT outputfile [DEPENDS dep1 [dep2...]])
-  function(generate_out)
-    set(options)
-    set(oneValueArgs INPUT OUTPUT)
-    set(multiValueArgs DEPENDS)
-    cmake_parse_arguments(_GO "${options}" "${oneValueArgs}" "${multiValueArgs}" ${ARGN})
-    if (NOT _GO_INPUT)
-      message(FATAL_ERROR "Invalid arguments.  generate_out requires input.")
-    endif()
-    if (NOT _GO_OUTPUT)
-      message(FATAL_ERROR "Invalid arguments.  generate_out requires output.")
-    endif()
-
-    add_custom_command(OUTPUT "${_GO_OUTPUT}"
-                       COMMAND "${CMAKE_COMMAND}"
-                               "-DINPUT=${_GO_INPUT}"
-                               "-DOUTPUT=${_GO_OUTPUT}"
-                               -P "${CMAKE_CURRENT_BINARY_DIR}/scripts/genout.cmake"
-                       DEPENDS "${_GO_INPUT}" ${_GO_DEPENDS}
-                       WORKING_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}")
-  endfunction()
-
-  # Generate specific source file with awk
-  # generate_source(OUTPUT outputfile [DEPENDS dep1 [dep2...]])
-  function(generate_source)
-    set(options)
-    set(oneValueArgs OUTPUT)
-    set(multiValueArgs DEPENDS)
-    cmake_parse_arguments(_GSO "${options}" "${oneValueArgs}" "${multiValueArgs}" ${ARGN})
-    if (NOT _GSO_OUTPUT)
-      message(FATAL_ERROR "Invalid arguments.  generate_source requires output.")
-    endif()
-
-    add_custom_command(OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/${_GSO_OUTPUT}"
-                       COMMAND "${CMAKE_COMMAND}"
-                               "-DOUTPUT=${_GSO_OUTPUT}"
-                               -P "${CMAKE_CURRENT_BINARY_DIR}/scripts/gensrc.cmake"
-                       DEPENDS ${_GSO_DEPENDS}
-                       WORKING_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}")
-  endfunction()
-
-  # Copy file
-  function(generate_copy source destination)
-    add_custom_command(OUTPUT "${destination}"
-                       COMMAND "${CMAKE_COMMAND}" -E remove "${destination}"
-                       COMMAND "${CMAKE_COMMAND}" -E copy "${source}"
-                                                          "${destination}"
-                       DEPENDS "${source}")
-  endfunction()
-
-  # Generate scripts/pnglibconf.h
-  generate_source(OUTPUT "scripts/pnglibconf.c"
-                  DEPENDS "${CMAKE_CURRENT_SOURCE_DIR}/scripts/pnglibconf.dfa"
-                          "${CMAKE_CURRENT_SOURCE_DIR}/scripts/options.awk"
-                          "${CMAKE_CURRENT_SOURCE_DIR}/pngconf.h")
-
-  # Generate pnglibconf.c
-  generate_source(OUTPUT "pnglibconf.c"
-                  DEPENDS "${CMAKE_CURRENT_SOURCE_DIR}/scripts/pnglibconf.dfa"
-                          "${CMAKE_CURRENT_SOURCE_DIR}/scripts/options.awk"
-                          "${CMAKE_CURRENT_SOURCE_DIR}/pngconf.h")
-
-  if(PNG_PREFIX)
-    set(PNGLIBCONF_H_EXTRA_DEPENDS
-        "${CMAKE_CURRENT_BINARY_DIR}/scripts/prefix.out"
-        "${CMAKE_CURRENT_SOURCE_DIR}/scripts/macro.lst")
-    set(PNGPREFIX_H_EXTRA_DEPENDS
-        "${CMAKE_CURRENT_BINARY_DIR}/scripts/intprefix.out")
-  endif()
-
-  generate_out(INPUT "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.c"
-               OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.out")
-
-  # Generate pnglibconf.h
-  generate_source(OUTPUT "pnglibconf.h"
-                  DEPENDS "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.out"
-                          ${PNGLIBCONF_H_EXTRA_DEPENDS})
-
-  generate_out(INPUT "${CMAKE_CURRENT_SOURCE_DIR}/scripts/intprefix.c"
-               OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/scripts/intprefix.out"
-               DEPENDS "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.h")
-
-  generate_out(INPUT "${CMAKE_CURRENT_SOURCE_DIR}/scripts/prefix.c"
-               OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/scripts/prefix.out"
-               DEPENDS "${CMAKE_CURRENT_SOURCE_DIR}/png.h"
-                       "${CMAKE_CURRENT_SOURCE_DIR}/pngconf.h"
-                       "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.out")
-
-  # Generate pngprefix.h
-  generate_source(OUTPUT "pngprefix.h"
-                  DEPENDS ${PNGPREFIX_H_EXTRA_DEPENDS})
-
-  generate_out(INPUT "${CMAKE_CURRENT_SOURCE_DIR}/scripts/sym.c"
-               OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/scripts/sym.out"
-               DEPENDS "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.h")
-
-  generate_out(INPUT "${CMAKE_CURRENT_SOURCE_DIR}/scripts/symbols.c"
-               OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/scripts/symbols.out"
-               DEPENDS "${CMAKE_CURRENT_SOURCE_DIR}/png.h"
-                       "${CMAKE_CURRENT_SOURCE_DIR}/pngconf.h"
-                       "${CMAKE_CURRENT_SOURCE_DIR}/scripts/pnglibconf.h.prebuilt")
-
-  generate_out(INPUT "${CMAKE_CURRENT_SOURCE_DIR}/scripts/vers.c"
-               OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/scripts/vers.out"
-               DEPENDS "${CMAKE_CURRENT_SOURCE_DIR}/png.h"
-                       "${CMAKE_CURRENT_SOURCE_DIR}/pngconf.h"
-                       "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.h")
-
-  generate_chk(INPUT "${CMAKE_CURRENT_BINARY_DIR}/scripts/symbols.out"
-               OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/scripts/symbols.chk"
-               DEPENDS "${CMAKE_CURRENT_SOURCE_DIR}/scripts/checksym.awk"
-                       "${CMAKE_CURRENT_SOURCE_DIR}/scripts/symbols.def")
-
-  add_custom_target(symbol-check DEPENDS
-                    "${CMAKE_CURRENT_BINARY_DIR}/scripts/symbols.chk")
-
-  generate_copy("${CMAKE_CURRENT_BINARY_DIR}/scripts/sym.out"
-                "${CMAKE_CURRENT_BINARY_DIR}/libpng.sym")
-  generate_copy("${CMAKE_CURRENT_BINARY_DIR}/scripts/vers.out"
-                "${CMAKE_CURRENT_BINARY_DIR}/libpng.vers")
-
-  add_custom_target(genvers DEPENDS "${CMAKE_CURRENT_BINARY_DIR}/libpng.vers")
-  add_custom_target(gensym DEPENDS "${CMAKE_CURRENT_BINARY_DIR}/libpng.sym")
-
-  add_custom_target("genprebuilt"
-                    COMMAND "${CMAKE_COMMAND}"
-                            "-DOUTPUT=scripts/pnglibconf.h.prebuilt"
-                            -P "${CMAKE_CURRENT_BINARY_DIR}/scripts/gensrc.cmake"
-                    WORKING_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}")
-
-  # A single target handles generation of all generated files.  If
-  # they are dependend upon separately by multiple targets, this
-  # confuses parallel make (it would require a separate top-level
-  # target for each file to track the dependencies properly).
-  add_custom_target(genfiles DEPENDS
-    "${CMAKE_CURRENT_BINARY_DIR}/libpng.sym"
-    "${CMAKE_CURRENT_BINARY_DIR}/libpng.vers"
-    "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.c"
-    "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.h"
-    "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.out"
-    "${CMAKE_CURRENT_BINARY_DIR}/pngprefix.h"
-    "${CMAKE_CURRENT_BINARY_DIR}/scripts/intprefix.out"
-    "${CMAKE_CURRENT_BINARY_DIR}/scripts/pnglibconf.c"
-    "${CMAKE_CURRENT_BINARY_DIR}/scripts/prefix.out"
-    "${CMAKE_CURRENT_BINARY_DIR}/scripts/sym.out"
-    "${CMAKE_CURRENT_BINARY_DIR}/scripts/symbols.chk"
-    "${CMAKE_CURRENT_BINARY_DIR}/scripts/symbols.out"
-    "${CMAKE_CURRENT_BINARY_DIR}/scripts/vers.out")
-endif(NOT AWK)
-
 # OUR SOURCES
 set(libpng_public_hdrs
   png.h
   pngconf.h
-  "${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.h"
+  ${CMAKE_CURRENT_BINARY_DIR}/pnglibconf.h
 )
-set(libpng_private_hdrs
-  pngpriv.h
-  pngdebug.h
-  pnginfo.h
-  pngstruct.h
-)
-if(AWK)
-  list(APPEND libpng_private_hdrs "${CMAKE_CURRENT_BINARY_DIR}/pngprefix.h")
-endif()
 set(libpng_sources
   ${libpng_public_hdrs}
-  ${libpng_private_hdrs}
+  pngdebug.h
+  pnginfo.h
+  pngpriv.h
+  pngstruct.h
   png.c
   pngerror.c
   pngget.c
@@ -431,8 +88,6 @@ set(libpng_sources
   pngwrite.c
   pngwtran.c
   pngwutil.c
-  ${libpng_arm_sources}
-  ${libpng_powerpc_sources}
 )
 set(pngtest_sources
   pngtest.c
@@ -443,18 +98,7 @@ set(pngvalid_sources
 set(pngstest_sources
   contrib/libtests/pngstest.c
 )
-set(pngunknown_sources
-  contrib/libtests/pngunknown.c
-)
-set(pngimage_sources
-  contrib/libtests/pngimage.c
-)
-set(pngfix_sources
-  contrib/tools/pngfix.c
-)
-set(png_fix_itxt_sources
-  contrib/tools/png-fix-itxt.c
-)
+# SOME NEEDED DEFINITIONS
 
 if(MSVC)
   add_definitions(-D_CRT_SECURE_NO_DEPRECATE)
@@ -470,59 +114,33 @@ include_directories(${CMAKE_CURRENT_SOURCE_DIR} ${ZLIB_INCLUDE_DIR})
 unset(PNG_LIB_TARGETS)
 
 if(PNG_SHARED)
-  add_library(png SHARED ${libpng_sources})
-  set(PNG_LIB_TARGETS png)
-  set_target_properties(png PROPERTIES OUTPUT_NAME ${PNG_LIB_NAME})
-  add_dependencies(png genfiles)
+  add_library(${PNG_LIB_NAME} SHARED ${libpng_sources})
+  set(PNG_LIB_TARGETS ${PNG_LIB_NAME})
   if(MSVC)
     # msvc does not append 'lib' - do it here to have consistent name
-    set_target_properties(png PROPERTIES PREFIX "lib")
-    set_target_properties(png PROPERTIES IMPORT_PREFIX "lib")
-  endif()
-  target_link_libraries(png ${ZLIB_LIBRARY} ${M_LIBRARY})
-
-  if(UNIX AND AWK)
-    if(HAVE_LD_VERSION_SCRIPT)
-      set_target_properties(png PROPERTIES LINK_FLAGS
-        "-Wl,--version-script='${CMAKE_CURRENT_BINARY_DIR}/libpng.vers'")
-    elseif(HAVE_SOLARIS_LD_VERSION_SCRIPT)
-      set_target_properties(png PROPERTIES LINK_FLAGS
-        "-Wl,-M -Wl,'${CMAKE_CURRENT_BINARY_DIR}/libpng.vers'")
-    endif()
+    set_target_properties(${PNG_LIB_NAME} PROPERTIES PREFIX "lib")
+    set_target_properties(${PNG_LIB_NAME} PROPERTIES IMPORT_PREFIX "lib")
   endif()
+  target_link_libraries(${PNG_LIB_NAME} ${ZLIB_LIBRARY} ${M_LIBRARY})
 endif()
 
 if(PNG_STATIC)
   # does not work without changing name
-  set(PNG_LIB_NAME_STATIC png_static)
-  add_library(png_static STATIC ${libpng_sources})
-  add_dependencies(png_static genfiles)
-  # MSVC doesn't use a different file extension for shared vs. static
-  # libs.  We are able to change OUTPUT_NAME to remove the _static
-  # for all other platforms.
-  if(NOT MSVC)
-    set_target_properties(png_static PROPERTIES
-      OUTPUT_NAME "${PNG_LIB_NAME}"
-      CLEAN_DIRECT_OUTPUT 1)
-  else()
-    set_target_properties(png_static PROPERTIES
-      OUTPUT_NAME "${PNG_LIB_NAME}_static"
-      CLEAN_DIRECT_OUTPUT 1)
-  endif()
-  list(APPEND PNG_LIB_TARGETS png_static)
+  set(PNG_LIB_NAME_STATIC ${PNG_LIB_NAME}_static)
+  add_library(${PNG_LIB_NAME_STATIC} STATIC ${libpng_sources})
+  list(APPEND PNG_LIB_TARGETS ${PNG_LIB_NAME_STATIC})
   if(MSVC)
     # msvc does not append 'lib' - do it here to have consistent name
-    set_target_properties(png_static PROPERTIES PREFIX "lib")
+    set_target_properties(${PNG_LIB_NAME_STATIC} PROPERTIES PREFIX "lib")
   endif()
-  target_link_libraries(png_static ${ZLIB_LIBRARY} ${M_LIBRARY})
+  target_link_libraries(${PNG_LIB_NAME_STATIC} ${ZLIB_LIBRARY} ${M_LIBRARY})
 endif()
 
 if(PNG_FRAMEWORK)
-  set(PNG_LIB_NAME_FRAMEWORK png_framework)
-  add_library(png_framework SHARED ${libpng_sources})
-  add_dependencies(png_framework genfiles)
-  list(APPEND PNG_LIB_TARGETS png_framework)
-  set_target_properties(png_framework PROPERTIES
+  set(PNG_LIB_NAME_FRAMEWORK ${PNG_LIB_NAME}_framework)
+  add_library(${PNG_LIB_NAME_FRAMEWORK} SHARED ${libpng_sources})
+  list(APPEND PNG_LIB_TARGETS ${PNG_LIB_NAME_FRAMEWORK})
+  set_target_properties(${PNG_LIB_NAME_FRAMEWORK} PROPERTIES
     FRAMEWORK TRUE
     FRAMEWORK_VERSION ${PNGLIB_VERSION}
     MACOSX_FRAMEWORK_SHORT_VERSION_STRING ${PNGLIB_MAJOR}.${PNGLIB_MINOR}
@@ -531,7 +149,7 @@ if(PNG_FRAMEWORK)
     XCODE_ATTRIBUTE_INSTALL_PATH "@rpath"
     PUBLIC_HEADER "${libpng_public_hdrs}"
     OUTPUT_NAME png)
-  target_link_libraries(png_framework ${ZLIB_LIBRARY} ${M_LIBRARY})
+  target_link_libraries(${PNG_LIB_NAME_FRAMEWORK} ${ZLIB_LIBRARY} ${M_LIBRARY})
 endif()
 
 if(NOT PNG_LIB_TARGETS)
@@ -542,170 +160,52 @@ if(NOT PNG_LIB_TARGETS)
 endif()
 
 if(PNG_SHARED AND WIN32)
-  set_target_properties(png PROPERTIES DEFINE_SYMBOL PNG_BUILD_DLL)
+  set_target_properties(${PNG_LIB_NAME} PROPERTIES DEFINE_SYMBOL PNG_BUILD_DLL)
 endif()
 
-function(png_add_test)
-  set(options)
-  set(oneValueArgs NAME COMMAND)
-  set(multiValueArgs OPTIONS FILES)
-  cmake_parse_arguments(_PAT "${options}" "${oneValueArgs}" "${multiValueArgs}" ${ARGN})
-
-  if (NOT _PAT_NAME)
-    message(FATAL_ERROR "Invalid arguments.  png_add_test requires name.")
-  endif()
-  if (NOT _PAT_COMMAND)
-    message(FATAL_ERROR "Invalid arguments.  png_add_test requires command.")
-  endif()
-
-  set(TEST_OPTIONS "${_PAT_OPTIONS}")
-  set(TEST_FILES "${_PAT_FILES}")
-
-  configure_file("${CMAKE_CURRENT_SOURCE_DIR}/scripts/test.cmake.in"
-                 "${CMAKE_CURRENT_BINARY_DIR}/tests/${_PAT_NAME}.cmake" @ONLY)
-  if(CMAKE_MAJOR_VERSION GREATER 2) # have generator expressions
-    add_test(NAME "${_PAT_NAME}"
-             COMMAND "${CMAKE_COMMAND}"
-             "-DLIBPNG=$<TARGET_FILE:png>"
-             "-DTEST_COMMAND=$<TARGET_FILE:${_PAT_COMMAND}>"
-             -P "${CMAKE_CURRENT_BINARY_DIR}/tests/${_PAT_NAME}.cmake")
-  else() # old 2.x add_test; limited and won't work well on Windows
-    # Note LIBPNG is a dummy value as there are no generator expressions
-    add_test("${_PAT_NAME}" "${CMAKE_COMMAND}"
-             "-DLIBPNG=${CMAKE_CURRENT_BINARY_DIR}/libpng.so"
-             "-DTEST_COMMAND=./${_PAT_COMMAND}"
-             -P "${CMAKE_CURRENT_BINARY_DIR}/tests/${_PAT_NAME}.cmake")
-  endif()
-endfunction()
-
 if(PNG_TESTS AND PNG_SHARED)
-  # Find test PNG files by globbing, but sort lists to ensure
-  # consistency between different filesystems.
-  file(GLOB PNGSUITE_PNGS "${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/*.png")
-  list(SORT PNGSUITE_PNGS)
-  file(GLOB TEST_PNGS "${CMAKE_CURRENT_SOURCE_DIR}/contrib/testpngs/*.png")
-  list(SORT TEST_PNGS)
-
-  set(PNGTEST_PNG "${CMAKE_CURRENT_SOURCE_DIR}/pngtest.png")
-
+  # does not work with msvc due to png_lib_ver issue
   add_executable(pngtest ${pngtest_sources})
-  target_link_libraries(pngtest png)
-
-  png_add_test(NAME pngtest COMMAND pngtest FILES "${PNGTEST_PNG}")
-
+  target_link_libraries(pngtest ${PNG_LIB_NAME})
+  add_test(pngtest ./pngtest ${CMAKE_CURRENT_SOURCE_DIR}/pngtest.png)
+  #
   add_executable(pngvalid ${pngvalid_sources})
-  target_link_libraries(pngvalid png)
-
-  png_add_test(NAME pngvalid-gamma-16-to-8
-               COMMAND pngvalid OPTIONS --gamma-16-to-8)
-  png_add_test(NAME pngvalid-gamma-alpha-mode
-               COMMAND pngvalid OPTIONS --gamma-alpha-mode)
-  png_add_test(NAME pngvalid-gamma-background
-               COMMAND pngvalid OPTIONS --gamma-background)
-  png_add_test(NAME pngvalid-gamma-expand16-alpha-mode
-               COMMAND pngvalid OPTIONS --gamma-alpha-mode --expand16)
-  png_add_test(NAME pngvalid-gamma-expand16-background
-               COMMAND pngvalid OPTIONS --gamma-background --expand16)
-  png_add_test(NAME pngvalid-gamma-expand16-transform
-               COMMAND pngvalid OPTIONS --gamma-transform --expand16)
-  png_add_test(NAME pngvalid-gamma-sbit
-               COMMAND pngvalid OPTIONS --gamma-sbit)
-  png_add_test(NAME pngvalid-gamma-threshold
-               COMMAND pngvalid OPTIONS --gamma-threshold)
-  png_add_test(NAME pngvalid-gamma-transform
-               COMMAND pngvalid OPTIONS --gamma-transform)
-  png_add_test(NAME pngvalid-progressive-interlace-standard
-               COMMAND pngvalid OPTIONS --standard --progressive-read --interlace)
-  png_add_test(NAME pngvalid-progressive-size
-               COMMAND pngvalid OPTIONS --size --progressive-read)
-  png_add_test(NAME pngvalid-progressive-standard
-               COMMAND pngvalid OPTIONS --standard --progressive-read)
-  png_add_test(NAME pngvalid-standard
-               COMMAND pngvalid OPTIONS --standard)
-  png_add_test(NAME pngvalid-transform
-               COMMAND pngvalid OPTIONS --transform)
-
+  target_link_libraries(pngvalid ${PNG_LIB_NAME})
+  add_test(pngvalid ./pngvalid)
   add_executable(pngstest ${pngstest_sources})
-  target_link_libraries(pngstest png)
-
-  foreach(gamma_type 1.8 linear none sRGB)
-    foreach(alpha_type none alpha)
-      set(PNGSTEST_FILES)
-      foreach(test_png ${TEST_PNGS})
-        string(REGEX MATCH ".*-linear[-.].*" TEST_PNG_LINEAR "${test_png}")
-        string(REGEX MATCH ".*-sRGB[-.].*" TEST_PNG_SRGB "${test_png}")
-        string(REGEX MATCH ".*-1.8[-.].*" TEST_PNG_G18 "${test_png}")
-        string(REGEX MATCH ".*-alpha-.*" TEST_PNG_ALPHA "${test_png}")
-
-        set(TEST_PNG_VALID TRUE)
-
-        if(TEST_PNG_ALPHA)
-          if (NOT "${alpha_type}" STREQUAL "alpha")
-            set(TEST_PNG_VALID FALSE)
-          endif()
-        else()
-          if ("${alpha_type}" STREQUAL "alpha")
-            set(TEST_PNG_VALID FALSE)
-          endif()
-        endif()
-
-        if(TEST_PNG_LINEAR)
-          if(NOT "${gamma_type}" STREQUAL "linear")
-            set(TEST_PNG_VALID FALSE)
-          endif()
-        elseif(TEST_PNG_SRGB)
-          if(NOT "${gamma_type}" STREQUAL "sRGB")
-            set(TEST_PNG_VALID FALSE)
-          endif()
-        elseif(TEST_PNG_G18)
-          if(NOT "${gamma_type}" STREQUAL "1.8")
-            set(TEST_PNG_VALID FALSE)
-          endif()
-        else()
-          if(NOT "${gamma_type}" STREQUAL "none")
-            set(TEST_PNG_VALID FALSE)
-          endif()
-        endif()
-
-        if(TEST_PNG_VALID)
-          list(APPEND PNGSTEST_FILES "${test_png}")
-        endif()
-      endforeach()
-      # Should already be sorted, but sort anyway to be certain.
-      list(SORT PNGSTEST_FILES)
-      png_add_test(NAME pngstest-${gamma_type}-${alpha_type}
-                   COMMAND pngstest
-                   OPTIONS --tmpfile "${gamma_type}-${alpha_type}-" --log
-                   FILES ${PNGSTEST_FILES})
-    endforeach()
-  endforeach()
-
-  add_executable(pngunknown ${pngunknown_sources})
-  target_link_libraries(pngunknown png)
-
-  png_add_test(NAME pngunknown-discard COMMAND pngunknown OPTIONS --strict default=discard FILES "${PNGTEST_PNG}")
-  png_add_test(NAME pngunknown-IDAT COMMAND pngunknown OPTIONS --strict default=discard IDAT=save FILES "${PNGTEST_PNG}")
-  png_add_test(NAME pngunknown-if-safe COMMAND pngunknown OPTIONS --strict default=if-safe FILES "${PNGTEST_PNG}")
-  png_add_test(NAME pngunknown-sAPI COMMAND pngunknown OPTIONS --strict bKGD=save cHRM=save gAMA=save all=discard iCCP=save sBIT=save sRGB=save FILES "${PNGTEST_PNG}")
-  png_add_test(NAME pngunknown-save COMMAND pngunknown OPTIONS --strict default=save FILES "${PNGTEST_PNG}")
-  png_add_test(NAME pngunknown-sTER COMMAND pngunknown OPTIONS --strict sTER=if-safe FILES "${PNGTEST_PNG}")
-  png_add_test(NAME pngunknown-vpAg COMMAND pngunknown OPTIONS --strict vpAg=if-safe FILES "${PNGTEST_PNG}")
-
-  add_executable(pngimage ${pngimage_sources})
-  target_link_libraries(pngimage png)
-
-  png_add_test(NAME pngimage-quick COMMAND pngimage OPTIONS --list-combos --log FILES ${PNGSUITE_PNGS})
-  png_add_test(NAME pngimage-full COMMAND pngimage OPTIONS --exhaustive --list-combos --log FILES ${PNGSUITE_PNGS})
-endif()
-
-if(PNG_SHARED)
-  add_executable(pngfix ${pngfix_sources})
-  target_link_libraries(pngfix png)
-  set(PNG_BIN_TARGETS pngfix)
-
-  add_executable(png-fix-itxt ${png_fix_itxt_sources})
-  target_link_libraries(png-fix-itxt ${ZLIB_LIBRARY} ${M_LIBRARY})
-  list(APPEND PNG_BIN_TARGETS png-fix-itxt)
+  target_link_libraries(pngstest ${PNG_LIB_NAME})
+  add_test(pngstest ./pngstest
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn0g01.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn0g02.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn0g04.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn0g08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn0g16.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn2c08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn2c16.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn3p01.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn3p02.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn3p04.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn3p08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn4a08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn4a16.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn6a08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/basn6a16.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbbn0g01.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbbn0g02.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbbn0g04.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbbn2c16.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbbn3p08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbgn2c16.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbgn3p08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbrn2c08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbwn0g16.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbwn3p08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftbyn3p08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftp0n0g08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftp0n2c08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftp0n3p08.png
+    ${CMAKE_CURRENT_SOURCE_DIR}/contrib/pngsuite/ftp1n3p08.png
+  )
 endif()
 
 # Ensure the CMAKE_LIBRARY_OUTPUT_DIRECTORY is set
@@ -734,15 +234,6 @@ macro(CREATE_SYMLINK SRC_FILE DEST_FILE)
   endif(WIN32 AND NOT CYGWIN AND NOT MSYS)
 endmacro()
 
-# Create source generation scripts.
-configure_file(${CMAKE_CURRENT_SOURCE_DIR}/scripts/genchk.cmake.in
-               ${CMAKE_CURRENT_BINARY_DIR}/scripts/genchk.cmake @ONLY)
-configure_file(${CMAKE_CURRENT_SOURCE_DIR}/scripts/genout.cmake.in
-               ${CMAKE_CURRENT_BINARY_DIR}/scripts/genout.cmake @ONLY)
-configure_file(${CMAKE_CURRENT_SOURCE_DIR}/scripts/gensrc.cmake.in
-               ${CMAKE_CURRENT_BINARY_DIR}/scripts/gensrc.cmake @ONLY)
-
-
 # libpng is a library so default to 'lib'
 if(NOT DEFINED CMAKE_INSTALL_LIBDIR)
   set(CMAKE_INSTALL_LIBDIR lib)
@@ -769,12 +260,22 @@ endif(NOT WIN32 OR CYGWIN OR MINGW)
 
 # SET UP LINKS
 if(PNG_SHARED)
-  set_target_properties(png PROPERTIES
-#   VERSION 16.${PNGLIB_RELEASE}.1.6.30beta02
-    VERSION 16.${PNGLIB_RELEASE}.0
-    SOVERSION 16
+  set_target_properties(${PNG_LIB_NAME} PROPERTIES
+#   VERSION 17.${PNGLIB_RELEASE}.1.7.0beta68
+    VERSION 17.${PNGLIB_RELEASE}.0
+    SOVERSION 17
     CLEAN_DIRECT_OUTPUT 1)
 endif()
+if(PNG_STATIC)
+  # MSVC doesn't use a different file extension for shared vs. static
+  # libs.  We are able to change OUTPUT_NAME to remove the _static
+  # for all other platforms.
+  if(NOT MSVC)
+    set_target_properties(${PNG_LIB_NAME_STATIC} PROPERTIES
+      OUTPUT_NAME ${PNG_LIB_NAME}
+      CLEAN_DIRECT_OUTPUT 1)
+  endif()
+endif()
 
 # If CMake > 2.4.x, we set a variable used below to export
 # targets to an export file.
@@ -795,16 +296,16 @@ if(NOT SKIP_INSTALL_LIBRARIES AND NOT SKIP_INSTALL_ALL )
       FRAMEWORK DESTINATION ${CMAKE_INSTALL_LIBDIR})
 
   if(PNG_SHARED)
-    # Create a symlink for libpng.dll.a => libpng16.dll.a on Cygwin
+    # Create a symlink for libpng.dll.a => libpng17.dll.a on Cygwin
     if(CYGWIN OR MINGW)
-       get_target_property(BUILD_TARGET_LOCATION png LOCATION_${CMAKE_BUILD_TYPE})
+       get_target_property(BUILD_TARGET_LOCATION ${PNG_LIB_NAME} LOCATION_${CMAKE_BUILD_TYPE})
        CREATE_SYMLINK(${BUILD_TARGET_LOCATION} libpng${CMAKE_IMPORT_LIBRARY_SUFFIX})
        install(FILES ${CMAKE_CURRENT_BINARY_DIR}/libpng${CMAKE_IMPORT_LIBRARY_SUFFIX}
          DESTINATION ${CMAKE_INSTALL_LIBDIR})
     endif(CYGWIN OR MINGW)
 
     if(NOT WIN32)
-      get_target_property(BUILD_TARGET_LOCATION png LOCATION_${CMAKE_BUILD_TYPE})
+      get_target_property(BUILD_TARGET_LOCATION ${PNG_LIB_NAME} LOCATION_${CMAKE_BUILD_TYPE})
       CREATE_SYMLINK(${BUILD_TARGET_LOCATION} libpng${CMAKE_SHARED_LIBRARY_SUFFIX})
       install(FILES ${CMAKE_CURRENT_BINARY_DIR}/libpng${CMAKE_SHARED_LIBRARY_SUFFIX}
          DESTINATION ${CMAKE_INSTALL_LIBDIR})
@@ -813,7 +314,7 @@ if(NOT SKIP_INSTALL_LIBRARIES AND NOT SKIP_INSTALL_ALL )
 
   if(PNG_STATIC)
     if(NOT WIN32 OR CYGWIN OR MINGW)
-      get_target_property(BUILD_TARGET_LOCATION png_static LOCATION_${CMAKE_BUILD_TYPE})
+      get_target_property(BUILD_TARGET_LOCATION ${PNG_LIB_NAME_STATIC} LOCATION_${CMAKE_BUILD_TYPE})
       CREATE_SYMLINK(${BUILD_TARGET_LOCATION} libpng${CMAKE_STATIC_LIBRARY_SUFFIX})
       install(FILES ${CMAKE_CURRENT_BINARY_DIR}/libpng${CMAKE_STATIC_LIBRARY_SUFFIX}
          DESTINATION ${CMAKE_INSTALL_LIBDIR})
@@ -833,11 +334,6 @@ if(NOT SKIP_INSTALL_EXECUTABLES AND NOT SKIP_INSTALL_ALL )
   endif(NOT WIN32 OR CYGWIN OR MINGW)
 endif()
 
-if(NOT SKIP_INSTALL_PROGRAMS AND NOT SKIP_INSTALL_ALL )
-  install(TARGETS ${PNG_BIN_TARGETS}
-      RUNTIME DESTINATION bin)
-endif()
-
 if(NOT SKIP_INSTALL_FILES AND NOT SKIP_INSTALL_ALL )
   # Install man pages
   if(NOT PNG_MAN_DIR)
@@ -875,3 +371,4 @@ endif()
 # to create msvc import lib for mingw compiled shared lib
 # pexports libpng.dll > libpng.def
 # lib /def:libpng.def /machine:x86
+

INSTALL

@@ -1,25 +1,24 @@
 
-    Installing libpng
+Installing libpng
 
 Contents
 
-       I. Simple installation
-      II. Rebuilding the configure scripts
-     III. Using scripts/makefile*
-      IV. Using cmake
-       V. Directory structure
-      VI. Building with project files
-     VII. Building with makefiles
-    VIII. Configuring libpng for 16-bit platforms
-      IX. Configuring for DOS
-       X. Configuring for Medium Model
-      XI. Prepending a prefix to exported symbols
-     XII. Configuring for compiler xxx:
-    XIII. Removing unwanted object code
-     XIV. Changes to the build and configuration of libpng in libpng-1.5.x
-      XV. Setjmp/longjmp issues
-     XVI. Common linking failures
-    XVII. Other sources of information about libpng
+   I. Simple installation
+  II. Rebuilding the configure scripts
+ III. Using scripts/makefile*
+  IV. Using cmake
+   V. Directory structure
+  VI. Building with project files
+ VII. Building with makefiles
+VIII. Configuring libpng for 16-bit platforms
+  IX. Configuring for DOS
+   X. Configuring for Medium Model
+  XI. Prepending a prefix to exported symbols
+ XII. Configuring for compiler xxx:
+XIII. Removing unwanted object code
+ XIV. Changes to the build and configuration of libpng in libpng-1.5.x
+  XV. Setjmp/longjmp issues
+ XVI. Other sources of information about libpng
 
 I. Simple installation
 
@@ -48,9 +47,7 @@ If configure does not work on your system, or if you have a need to
 change configure.ac or Makefile.am, and you have a reasonably
 up-to-date set of tools, running ./autogen.sh in a git clone before
 running ./configure may fix the problem.  To be really sure that you
-aren't using any of the included pre-built scripts, especially if you
-are building from a tar distribution instead of a git distribution,
-do this:
+aren't using any of the included pre-built scripts, you can do this:
 
     ./configure --enable-maintainer-mode
     make maintainer-clean
@@ -79,7 +76,7 @@ Or you can use one of the "projects" in the "projects" directory.
 Before installing libpng, you must first install zlib, if it
 is not already on your system.  zlib can usually be found
 wherever you got libpng; otherwise go to http://zlib.net.  You can place
-zlib in the same directory as libpng or in another directory.
+zlib in in the same directory as libpng or in another directory.
 
 If your system already has a preinstalled zlib you will still need
 to have access to the zlib.h and zconf.h include files that
@@ -90,24 +87,22 @@ standard library search path, put ZLIBLIB, ZLIBINC, CPPFLAGS, LDFLAGS,
 and LD_LIBRARY_PATH in your environment before running "make test"
 or "make distcheck":
 
-    ZLIBLIB=/path/to/lib export ZLIBLIB
-    ZLIBINC=/path/to/include export ZLIBINC
-    CPPFLAGS="-I$ZLIBINC" export CPPFLAGS
-    LDFLAGS="-L$ZLIBLIB" export LDFLAGS
-    LD_LIBRARY_PATH="$ZLIBLIB:$LD_LIBRARY_PATH" export LD_LIBRARY_PATH
+ZLIBLIB=/path/to/lib export ZLIBLIB
+ZLIBINC=/path/to/include export ZLIBINC
+CPPFLAGS="-I$ZLIBINC" export CPPFLAGS
+LDFLAGS="-L$ZLIBLIB" export LDFLAGS
+LD_LIBRARY_PATH="$ZLIBLIB:$LD_LIBRARY_PATH" export LD_LIBRARY_PATH
 
 If you are using one of the makefile scripts, put ZLIBLIB and ZLIBINC
-in your environment and type
-
-    make ZLIBLIB=$ZLIBLIB ZLIBINC=$ZLIBINC test
+in your environment and type "make ZLIBLIB=$ZLIBLIB ZLIBINC=$ZLIBINC test".
 
 IV. Using cmake
 
 If you want to use "cmake" (see www.cmake.org), type
 
-    cmake . -DCMAKE_INSTALL_PREFIX=/path
-    make
-    make install
+   cmake . -DCMAKE_INSTALL_PREFIX=/path
+   make
+   make install
 
 As when using the simple configure method described above, "/path" points to
 the installation directory where you want to put the libpng "lib", "include",
@@ -121,8 +116,8 @@ or "zlib128") so that you have directories called "zlib" and "libpng".
 
 Your directory structure should look like this:
 
-    .. (the parent directory)
-      libpng (this directory)
+   ..       (the parent directory)
+      libpng  (this directory)
           INSTALL (this file)
           README
           *.h, *.c  => libpng source files
@@ -165,15 +160,10 @@ VII. Building with makefiles
 Copy the file (or files) that you need from the
 scripts directory into this directory, for example
 
-MSDOS example:
-
-    copy scripts\makefile.msc makefile
-    copy scripts\pnglibconf.h.prebuilt pnglibconf.h
-
-UNIX example:
-
-    cp scripts/makefile.std makefile
-    cp scripts/pnglibconf.h.prebuilt pnglibconf.h
+   MSDOS example: copy scripts\makefile.msc makefile
+                  copy scripts\pnglibconf.h.prebuilt pnglibconf.h
+   UNIX example:  cp scripts/makefile.std makefile
+                  cp scripts/pnglibconf.h.prebuilt pnglibconf.h
 
 Read the makefile to see if you need to change any source or
 target directories to match your preferences.
@@ -249,7 +239,7 @@ libpng are compiled.  All the defines end in _SUPPORTED.  If you are
 never going to use a capability, you can change the #define to #undef
 before recompiling libpng and save yourself code and data space, or
 you can turn off individual capabilities with defines that begin with
-"PNG_NO_".
+PNG_NO_.
 
 In libpng-1.5.0 and later, the #define's are in pnglibconf.h instead.
 
@@ -317,7 +307,7 @@ only png_longjmp_ptr, which must match the C longjmp function.)  The new
 approach is documented in pngconf.h
 
 Despite these changes, libpng 1.5.0 only supports the native C function
-calling standard on those platforms tested so far ("__cdecl" on Microsoft
+calling standard on those platforms tested so far (__cdecl on Microsoft
 Windows).  This is because the support requirements for alternative
 calling conventions seem to no longer exist.  Developers who find it
 necessary to set PNG_API_RULE to 1 should advise the mailing list
@@ -381,7 +371,7 @@ your application is going to be using multiple threads, you should
 configure libpng with PNG_NO_SETJMP in your pngusr.dfa file, with
 -DPNG_NO_SETJMP on your compile line, or with
 
-    #undef PNG_SETJMP_SUPPORTED
+  #undef PNG_SETJMP_SUPPORTED
 
 in your pnglibconf.h or pngusr.h.
 
@@ -390,25 +380,28 @@ This requires setjmp/longjmp, so you must either build the library
 with PNG_SETJMP_SUPPORTED defined, or with PNG_SIMPLIFIED_READ_SUPPORTED
 and PNG_SIMPLIFIED_WRITE_SUPPORTED undefined.
 
-XVI. Common linking failures
-
-If your application fails to find libpng or zlib entries while linking:
-
-  Be sure "-lz" appears after "-lpng" on your linking command.
-
-  Be sure you have built libpng, zlib, and your application for the
-  same platform (e.g., 32-bit or 64-bit).
-
-  If you are using the vstudio project, observe the WARNING in
-  project/vstudio/README.txt.
-
-XVII. Other sources of information about libpng:
+XVI. Other sources of information about libpng:
 
 Further information can be found in the README and libpng-manual.txt
 files, in the individual makefiles, in png.h, and the manual pages
 libpng.3 and png.5.
 
-Copyright (c) 1998-2002,2006-2016 Glenn Randers-Pehrson
-This document is released under the libpng license.
-For conditions of distribution and use, see the disclaimer
-and license in png.h.
+Using the ./configure script -- 16 December 2002.
+=================================================
+
+The ./configure script should work compatibly with what scripts/makefile.*
+did, however there are some options you might need to add to configure
+explicitly, which previously was done semi-automatically (if you didn't edit
+scripts/makefile.* yourself, that is)
+
+CFLAGS="-Wall -O -funroll-loops \
+-malign-loops=2 -malign-functions=2" ./configure --prefix=/usr/include \
+--with-pkgconfigdir=/usr/lib/pkgconfig --includedir=/usr/include
+
+You can alternatively specify --includedir=/usr/include, /usr/local/include,
+/usr/include/libpng17, or whatever.
+
+If you find that the configure script is out-of-date or is not supporting
+your platform properly, try running autogen.sh to regenerate "configure",
+"Makefile.in", and the other configuration files. Then try configure again.
+

LICENSE

@@ -10,8 +10,8 @@ this sentence.
 
 This code is released under the libpng license.
 
-libpng versions 1.0.7, July 1, 2000 through 1.6.30beta02, April 22, 2017 are
-Copyright (c) 2000-2002, 2004, 2006-2017 Glenn Randers-Pehrson, are
+libpng versions 1.0.7, July 1, 2000, through 1.7.0beta68, November 12, 2015, are
+Copyright (c) 2000-2002, 2004, 2006-2015 Glenn Randers-Pehrson, are
 derived from libpng-1.0.6, and are distributed according to the same
 disclaimer and license as libpng-1.0.6 with the following individuals
 added to the list of Contributing Authors:
@@ -22,9 +22,6 @@ added to the list of Contributing Authors:
    Cosmin Truta
    Gilles Vollant
    James Yu
-   Mandar Sahastrabuddhe
-   Google Inc.
-   Vadim Barkov
 
 and with the following additions to the disclaimer:
 
@@ -35,10 +32,6 @@ and with the following additions to the disclaimer:
    risk of satisfactory quality, performance, accuracy, and effort is with
    the user.
 
-Some files in the "contrib" directory and some configure-generated
-files that are distributed with libpng have other copyright owners and
-are released under other open source licenses.
-
 libpng versions 0.97, January 1998, through 1.0.6, March 20, 2000, are
 Copyright (c) 1998-2000 Glenn Randers-Pehrson, are derived from
 libpng-0.96, and are distributed according to the same disclaimer and
@@ -62,9 +55,6 @@ Contributing Authors:
    Greg Roelofs
    Tom Tanner
 
-Some files in the "scripts" directory have other copyright owners
-but are released under this license.
-
 libpng versions 0.5, May 1995, through 0.88, January 1996, are
 Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc.
 
@@ -105,29 +95,18 @@ appreciated.
 
 END OF COPYRIGHT NOTICE, DISCLAIMER, and LICENSE.
 
-TRADEMARK:
+A "png_get_copyright" function is available, for convenient use in "about"
+boxes and the like:
 
-The name "libpng" has not been registered by the Copyright owner
-as a trademark in any jurisdiction.  However, because libpng has
-been distributed and maintained world-wide, continually since 1995,
-the Copyright owner claims "common-law trademark protection" in any
-jurisdiction where common-law trademark is recognized.
+   printf("%s", png_get_copyright(NULL));
 
-OSI CERTIFICATION:
+Also, the PNG logo (in PNG format, of course) is supplied in the
+files "pngbar.png" and "pngbar.jpg (88x31) and "pngnow.png" (98x31).
 
 Libpng is OSI Certified Open Source Software.  OSI Certified Open Source is
 a certification mark of the Open Source Initiative. OSI has not addressed
 the additional disclaimers inserted at version 1.0.7.
 
-EXPORT CONTROL:
-
-The Copyright owner believes that the Export Control Classification
-Number (ECCN) for libpng is EAR99, which means not subject to export
-controls or International Traffic in Arms Regulations (ITAR) because
-it is open source, publicly available software, that does not contain
-any encryption software.  See the EAR, paragraphs 734.3(b)(3) and
-734.7(b).
-
 Glenn Randers-Pehrson
 glennrp at users.sourceforge.net
-April 22, 2017
+November 12, 2015

Makefile.am

@@ -1,21 +1,13 @@
-# Makefile.am, the source file for Makefile.in (and hence Makefile), is
+# Makefile.am:
+#   Source file for Makefile.in (and hence Makefile)
 #
-# Copyright (c) 2004-2016 Glenn Randers-Pehrson
-# Last changed in libpng 1.6.25 [September 1, 2016]
-#
-# This code is released under the libpng license.
-# For conditions of distribution and use, see the disclaimer
-# and license in png.h
 
 PNGLIB_BASENAME= libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@
 
 ACLOCAL_AMFLAGS = -I scripts
 
 # test programs - run on make check, make distcheck
-check_PROGRAMS= pngtest pngunknown pngstest pngvalid pngimage pngcp
-if HAVE_CLOCK_GETTIME
-check_PROGRAMS += timepng
-endif
+check_PROGRAMS= pngtest pngunknown pngstest pngvalid pngimage
 
 # Utilities - installed
 bin_PROGRAMS= pngfix png-fix-itxt
@@ -45,17 +37,11 @@ pngunknown_LDADD = libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@.la
 pngimage_SOURCES = contrib/libtests/pngimage.c
 pngimage_LDADD = libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@.la
 
-timepng_SOURCES = contrib/libtests/timepng.c
-timepng_LDADD = libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@.la
-
 pngfix_SOURCES = contrib/tools/pngfix.c
 pngfix_LDADD = libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@.la
 
 png_fix_itxt_SOURCES = contrib/tools/png-fix-itxt.c
 
-pngcp_SOURCES = contrib/tools/pngcp.c
-pngcp_LDADD = libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@.la
-
 # Generally these are single line shell scripts to run a test with a particular
 # set of parameters:
 TESTS =\
@@ -90,26 +76,15 @@ lib_LTLIBRARIES=libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@.la
 libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES = png.c pngerror.c\
 	pngget.c pngmem.c pngpread.c pngread.c pngrio.c pngrtran.c pngrutil.c\
 	pngset.c pngtrans.c pngwio.c pngwrite.c pngwtran.c pngwutil.c\
-	png.h pngconf.h pngdebug.h pnginfo.h pngpriv.h pngstruct.h pngusr.dfa
+	png.h pngconf.h pngdebug.h pnginfo.h pngpriv.h pngstruct.h pngchunk.h \
+	pngusr.dfa
 
-if PNG_ARM_NEON
-libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES += arm/arm_init.c\
-	arm/filter_neon.S arm/filter_neon_intrinsics.c
-endif
-
-if PNG_MIPS_MSA
-libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES += mips/mips_init.c\
-	mips/filter_msa_intrinsics.c
-endif
-
-if PNG_INTEL_SSE
-libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES += intel/intel_init.c\
-	intel/filter_sse2_intrinsics.c
-endif
-
-if PNG_POWERPC_VSX
-libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES += powerpc/powerpc_init.c\
-        powerpc/filter_vsx_intrinsics.c
+# Include libpng extensions, if appropriate.  This uses a Makefile.am fragment,
+# pre-set things that might be changed by addition:
+AM_CPPFLAGS =
+DFA_EXTENSION = 
+if LIBPNG_EXTENSIONS
+include $(top_srcdir)/contrib/extensions.am
 endif
 
 nodist_libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES = pnglibconf.h
@@ -152,12 +127,12 @@ EXTRA_DIST= \
 	$(TESTS) $(XFAIL_TESTS) tests/pngstest \
 	CMakeLists.txt example.c libpng-manual.txt
 
-SCRIPT_CLEANFILES=scripts/*.out scripts/*.chk
+SCRIPT_CLEANFILES=scripts/*.out scripts/*.chk scripts/pnglibconf.c
 
 CLEANFILES= *.tf? pngout.png libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@.pc \
 	libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@-config libpng.vers libpng.sym \
 	check.new pnglibconf.h pngprefix.h symbols.new pngtest-log.txt \
-	pnglibconf.out pnglibconf.c pnglibconf.pre pnglibconf.dfn \
+	pnglibconf.out pnglibconf.c pnglibconf.dfn pnglibconf.pre \
 	$(SCRIPT_CLEANFILES)
 
 MAINTAINERCLEANFILES = Makefile.in aclocal.m4 config.guess config.h.in \
@@ -169,8 +144,8 @@ config.sub configure depcomp install-sh ltmain.sh missing
 PNG_COPTS = @PNG_COPTS@
 AM_CFLAGS = ${PNG_COPTS}
 
-# DFNCPP is normally just CPP - the C preprocessor - but on Solaris and maybe
-# other operating systems (NeXT?) the C preprocessor selected by configure
+# DFNCPP is normally just "CPP -E" - the C preprocessor - but on Solaris and
+# maybe other operating systems (NeXT?) the C preprocessor selected by configure
 # checks input tokens for validity - effectively it performs part of the ANSI-C
 # parsing - and therefore fails with the .df files.  configure.ac has special
 # checks for this and sets DFNCPP appropriately.
@@ -189,6 +164,8 @@ scripts/prefix.out: png.h pngconf.h pnglibconf.out
 scripts/symbols.out: png.h pngconf.h $(srcdir)/scripts/pnglibconf.h.prebuilt
 scripts/intprefix.out: pnglibconf.h
 
+pnglibconf.out: config.h
+
 libpng.sym: scripts/sym.out
 	rm -f $@
 	cp $? $@
@@ -200,7 +177,7 @@ if DO_PNG_PREFIX
 # Rename functions in scripts/prefix.out with a PNG_PREFIX prefix.
 # Rename macros in scripts/macro.lst from PNG_PREFIXpng_ to PNG_ (the actual
 # implementation of the macro).
-pnglibconf.h: pnglibconf.out scripts/prefix.out scripts/macro.lst
+pnglibconf.h: pnglibconf.out scripts/prefix.out scripts/macro.lst config.h
 	rm -f $@
 	$(AWK) 's==0 && NR>1{print prev}\
 	   s==0{prev=$$0}\
@@ -248,7 +225,6 @@ contrib/libtests/timepng.o: pnglibconf.h
 
 contrib/tools/makesRGB.o: pnglibconf.h
 contrib/tools/pngfix.o: pnglibconf.h
-contrib/tools/pngcp.o: pnglibconf.h
 
 # We must use -DPNG_NO_USE_READ_MACROS here even when the library may actually
 # be built with PNG_USE_READ_MACROS; this prevents the read macros from
@@ -266,17 +242,17 @@ endif
 	rm -f $@ $*.tf[12]
 	test -d scripts || mkdir scripts || test -d scripts
 	$(DFNCPP) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES)\
-	    $(CPPFLAGS) $(SYMBOL_CFLAGS) $< > $*.tf1
+	    $(AM_CPPFLAGS) $(CPPFLAGS) $(SYMBOL_CFLAGS) $< > $*.tf1
 	$(AWK) -f "${srcdir}/scripts/dfn.awk" out="$*.tf2" $*.tf1 1>&2
 	rm -f $*.tf1
 	mv $*.tf2 $@
 
 # The .c file for pnglibconf.h is machine generated
-pnglibconf.c: scripts/pnglibconf.dfa scripts/options.awk pngconf.h pngusr.dfa $(DFA_XTRA)
+pnglibconf.c: scripts/pnglibconf.dfa scripts/options.awk pngconf.h pngusr.dfa $(DFA_EXTENSION) $(DFA_XTRA)
 	rm -f $@ $*.tf[45]
 	$(AWK) -f ${srcdir}/scripts/options.awk out=$*.tf4 version=search\
 	    ${srcdir}/pngconf.h ${srcdir}/scripts/pnglibconf.dfa\
-	    ${srcdir}/pngusr.dfa $(DFA_XTRA) 1>&2
+	    $(DFA_EXTENSION) ${srcdir}/pngusr.dfa $(DFA_XTRA) 1>&2
 	$(AWK) -f ${srcdir}/scripts/options.awk out=$*.tf5 $*.tf4 1>&2
 	rm $*.tf4
 	mv $*.tf5 $@
@@ -294,8 +270,8 @@ scripts/symbols.chk: scripts/checksym.awk scripts/symbols.def scripts/symbols.ou
 # be empty - no non-standard defines
 scripts/pnglibconf.c: scripts/pnglibconf.dfa scripts/options.awk pngconf.h
 	rm -f $@ pnglibconf.tf[67]
-	test -z "$(CPPFLAGS)"
-	echo "com @PNGLIB_VERSION@ STANDARD API DEFINITION" |\
+	test -z "$(CPPFLAGS)" -a -z "$(AM_CPPFLAGS)"
+	echo "com libpng @PNGLIB_VERSION@ STANDARD API DEFINITION" |\
 	$(AWK) -f ${srcdir}/scripts/options.awk out=pnglibconf.tf6\
 	    logunsupported=1 version=search ${srcdir}/pngconf.h -\
 	    ${srcdir}/scripts/pnglibconf.dfa 1>&2
@@ -304,7 +280,7 @@ scripts/pnglibconf.c: scripts/pnglibconf.dfa scripts/options.awk pngconf.h
 	rm pnglibconf.tf6
 	mv pnglibconf.tf7 $@
 
-$(libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_OBJECTS): png.h pngconf.h \
+$(libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_OBJECTS): png.h pngconf.h pngchunk.h \
 	pnglibconf.h pngpriv.h pngdebug.h pnginfo.h pngstruct.h pngprefix.h
 
 test: check-am

README

@@ -1,4 +1,4 @@
-README for libpng version 1.6.30beta02 - April 22, 2017 (shared library 16.0)
+README for libpng version 1.7.0beta68 - November 12, 2015 (shared library 17.0)
 See the note about version numbers near the top of png.h
 
 See INSTALL for instructions on how to install libpng.
@@ -90,7 +90,7 @@ You may also want a copy of the PNG specification.  It is available
 as an RFC, a W3C Recommendation, and an ISO/IEC Standard.  You can find
 these at http://www.libpng.org/pub/png/documents/
 
-This code is currently being archived at libpng.sourceforge.io in the
+This code is currently being archived at libpng.sf.net in the
 [DOWNLOAD] area, and at ftp://ftp.simplesystems.org.  If you can't find it
 in any of those places, e-mail me, and I'll help you find it.
 
@@ -179,25 +179,18 @@ Files in this distribution:
       pngwtran.c    =>  Write data transformations
       pngwutil.c    =>  Write utility functions
       arm           =>  Contains optimized code for the ARM platform
-      powerpc       =>  Contains optimized code for the PowerPC platform
       contrib       =>  Contributions
-       arm-neon         =>  Optimized code for ARM-NEON platform
-       powerpc-vsx      =>  Optimized code for POWERPC-VSX platform
        examples         =>  Example programs
        gregbook         =>  source code for PNG reading and writing, from
                             Greg Roelofs' "PNG: The Definitive Guide",
                             O'Reilly, 1999
        libtests         =>  Test programs
-       mips-msa         =>  Optimized code for MIPS-MSA platform
        pngminim         =>  Minimal decoder, encoder, and progressive decoder
                             programs demonstrating use of pngusr.dfa
        pngminus         =>  Simple pnm2png and png2pnm programs
        pngsuite         =>  Test images
-       testpngs
        tools            =>  Various tools
        visupng          =>  Contains a MSVC workspace for VisualPng
-      intel             =>  Optimized code for INTEL-SSE2 platform
-      mips              =>  Optimized code for MIPS platform
       projects      =>  Contains project files and workspaces for
                         building a DLL
        owatcom          =>  Contains a WATCOM project for building libpng

TODO

@@ -5,14 +5,13 @@ Final bug fixes.
 Better C++ wrapper/full C++ implementation?
 Fix problem with C++ and EXTERN "C".
 cHRM transformation.
-Remove setjmp/longjmp usage in favor of returning error codes. As a start on
-  this, minimize the use of png_error(), replacing them with
-  png_warning(); return(0); or similar.
+Remove setjmp/longjmp usage in favor of returning error codes.
 Palette creation.
 Add "grayscale->palette" transformation and "palette->grayscale" detection.
 Improved dithering.
 Multi-lingual error and warning message support.
 Complete sRGB transformation (presently it simply uses gamma=0.45455).
+Make profile checking optional via a png_set_something() call.
 Man pages for function calls.
 Better documentation.
 Better filter selection
@@ -24,6 +23,7 @@ Use greater precision when changing to linear gamma for compositing against
   background and doing rgb-to-gray transformation.
 Investigate pre-incremented loop counters and other loop constructions.
 Add interpolated method of handling interlacing.
+Switch to the simpler zlib (zlib/libpng) license if legally possible.
 Extend pngvalid.c to validate more of the libpng transformations.
 Refactor preprocessor conditionals to compile entire statements
 

configure.ac

@@ -1,12 +1,5 @@
 # configure.ac
 
-# Copyright (c) 2004-2016 Glenn Randers-Pehrson
-# Last changed in libpng 1.6.25 [September 1, 2016]
-
-# This code is released under the libpng license.
-# For conditions of distribution and use, see the disclaimer
-# and license in png.h
-
 dnl Process this file with autoconf to produce a configure script.
 dnl
 dnl Minor upgrades (compatible ABI): increment the package version
@@ -25,7 +18,7 @@ AC_PREREQ([2.68])
 
 dnl Version number stuff here:
 
-AC_INIT([libpng],[1.6.30beta02],[png-mng-implement@lists.sourceforge.net])
+AC_INIT([libpng],[1.7.0beta68],[png-mng-implement@lists.sourceforge.net])
 AC_CONFIG_MACRO_DIR([scripts])
 
 # libpng does not follow GNU file name conventions (hence 'foreign')
@@ -34,9 +27,10 @@ AC_CONFIG_MACRO_DIR([scripts])
 # dist-xz requires automake 1.11 or later
 # 1.12.2 fixes a security issue in 1.11.2 and 1.12.1
 # 1.13 is required for parallel tests
-AM_INIT_AUTOMAKE([1.13 foreign dist-xz color-tests silent-rules subdir-objects])
+# 1.14.1 is required for %reldir% support
+AM_INIT_AUTOMAKE([1.15 foreign dist-xz color-tests silent-rules subdir-objects])
 # The following line causes --disable-maintainer-mode to be the default to
-# configure. This is necessary because libpng distributions cannot rely on the
+# configure, this is necessary because libpng distributions cannot rely on the
 # time stamps of the autotools generated files being correct
 AM_MAINTAINER_MODE
 
@@ -46,10 +40,10 @@ dnl automake, so the following is not necessary (and is not defined anyway):
 dnl AM_PREREQ([1.11.2])
 dnl stop configure from automagically running automake
 
-PNGLIB_VERSION=1.6.30beta02
+PNGLIB_VERSION=1.7.0beta68
 PNGLIB_MAJOR=1
-PNGLIB_MINOR=6
-PNGLIB_RELEASE=28
+PNGLIB_MINOR=7
+PNGLIB_RELEASE=0
 
 dnl End of version number stuff
 
@@ -67,10 +61,10 @@ AC_PROG_INSTALL
 AC_PROG_LN_S
 AC_PROG_MAKE_SET
 
-dnl libtool/libtoolize; version 2.4.2 is the tested version. This or any
+dnl libtool/libtoolize; version 2.4.4 is the tested version, this or any
 dnl compatible later version may be used
 LT_INIT([win32-dll])
-LT_PREREQ([2.4.2])
+LT_PREREQ([2.4.4])
 
 # Some awks crash when confronted with pnglibconf.dfa, do a test run now
 # to make sure this doesn't happen
@@ -86,7 +80,7 @@ fi
 
 # This is a remnant of the old cc -E validation, where it may have been
 # necessary to use a different preprocessor for .dfn files
-DFNCPP="$CPP"
+DFNCPP="$CPP -E"
 AC_SUBST(DFNCPP)
 
 # -Werror cannot be passed to GCC in CFLAGS because configure will fail (it
@@ -114,57 +108,148 @@ AC_ARG_ENABLE(werror,
       CFLAGS="$sav_CFLAGS"
     fi],)
 
-# For GCC 5 the default mode for C is -std=gnu11 instead of -std=gnu89
-# In pngpriv.h we request just the POSIX 1003.1 and C89 APIs by defining _POSIX_SOURCE to 1
-# This is incompatible with the new default mode, so we test for that and force the 
-# "-std=c89" compiler option:
-AC_MSG_CHECKING([if we need to force back C standard to C89])
-AC_COMPILE_IFELSE(
-   [AC_LANG_PROGRAM([
-      [#define _POSIX_SOURCE 1]
-      [#include <stdio.h>]
-   ])],
-   AC_MSG_RESULT(no),[
-      if test "x$GCC" != "xyes"; then
-         AC_MSG_ERROR(
-            [Forcing back to C89 is required but the flags are only known for GCC])
-      fi
-   AC_MSG_RESULT(yes)
-   CFLAGS="$CFLAGS -std=c89"
-])
-
 # Checks for header files.
 AC_HEADER_STDC
+# Headers used in pngpriv.h:
+AC_CHECK_HEADERS([assert.h stdlib.h string.h float.h fp.h math.h m68881.h])
+# Additional headers used in pngconf.h:
+AC_CHECK_HEADERS([limits.h stddef.h stdio.h setjmp.h time.h])
+# Used by pngread.c (for the simplified API):
+AC_CHECK_HEADERS([errno.h])
+# The following are not required for the standard 'make' build of libpng:
+# Additional requirements of contrib/libtests/makepng.c:
+AC_CHECK_HEADERS([ctype.h])
+# Additional requirements of contrib/libtests/pngvalid.c:
+AC_CHECK_HEADERS([fenv.h])
 
 # Checks for typedefs, structures, and compiler characteristics.
-AC_C_CONST
-AC_TYPE_SIZE_T
-AC_STRUCT_TM
+# Most of these have been removed from 1.7.0 because they checked for non-ANSI-C
+# behaviors and the result were, anyway, not used by earlier versions of libpng.
 AC_C_RESTRICT
 
-# Checks for library functions.
-AC_FUNC_STRTOD
-AC_CHECK_FUNCS([memset], , AC_MSG_ERROR(memset not found in libc))
-AC_CHECK_FUNCS([pow], , AC_CHECK_LIB(m, pow, , AC_MSG_ERROR(cannot find pow)) )
+# Checks for library functions; these need to be in order, least demanding to
+# most because the -l directives are added to the head of the list at each
+# check.
+# ANSI-C functions: all are required but they don't have to be implemented in
+# libc, it is perfectly ok to configure with an extra, wrapper, library preset
+# in LIBS.  NOTE: this list may not be complete
+AC_CHECK_FUNCS([memset memcpy strncpy],,
+               AC_MSG_FAILURE([missing ANSI-C functions]))
 
-# Some later POSIX 1003.1 functions are required for test programs, failure here
-# is soft (the corresponding test program is not built).
-AC_CHECK_FUNC([clock_gettime],,[AC_MSG_WARN([not building timepng])])
-AM_CONDITIONAL([HAVE_CLOCK_GETTIME], [test "$ac_cv_func_clock_gettime" = "yes"])
+# floor and ceil are required for FLOATING_POINT, but they can be turned off on
+# the configure command line (and they are independent.)  Default is to 'on'
+AC_ARG_ENABLE([floating-point],
+   AS_HELP_STRING([[[--enabled-floating-point]]],
+      [Enable floating point APIs (default: yes)]),,
+   [enable_floating_point=yes])
 
-AC_ARG_WITH(zlib-prefix,
-   AS_HELP_STRING([[[--with-zlib-prefix]]],
-      [prefix that may have been used in installed zlib]),
-      [ZPREFIX=${withval}],
-      [ZPREFIX='z_'])
-AC_CHECK_LIB(z, zlibVersion, ,
-    AC_CHECK_LIB(z, ${ZPREFIX}zlibVersion, , AC_MSG_ERROR(zlib not installed)))
+AC_ARG_ENABLE([fixed-point],
+   AS_HELP_STRING([[[--enable-fixed-point]]],
+      [Enable fixed point APIs (default: yes)]),,
+   [enable_fixed_point=yes])
+
+AC_ARG_ENABLE([floating-arithmetic],
+   AS_HELP_STRING([[[--enable-floating-arithmetic]]],
+      [Use floating point arithmetic internally (default: yes)]),,
+   [enable_floating_arithmetic=yes])
+
+# Check for a requirement for math library support
+if test "$enable_floating_arithmetic" = yes
+then
+   AC_SEARCH_LIBS([floor],[m],,
+                  [AC_MSG_WARN([not found])
+                   enable_floating_arithmetic=no])
+   AC_SEARCH_LIBS([frexp],[m],,
+                  [AC_MSG_WARN([not found])
+                   enable_floating_arithmetic=no])
+   AC_SEARCH_LIBS([modf],[m],,
+                  [AC_MSG_WARN([not found])
+                   enable_floating_arithmetic=no])
+   AC_SEARCH_LIBS([atof],[m],,
+                  [AC_MSG_WARN([not found])
+                   enable_floating_arithmetic=no])
+   AC_SEARCH_LIBS([pow],[m],,
+                  [AC_MSG_WARN([not found])
+                   enable_floating_arithmetic=no])
+fi
+
+# Set the relevant defines if required (only if required), this will override
+# anything on the command line.
+if test "$enable_floating_point" != yes
+then
+   AC_MSG_NOTICE([Floating point APIs disabled])
+   AC_DEFINE([PNG_NO_FLOATING_POINT],[1],
+             [No support for floating point APIs])
+fi
+if test "$enable_fixed_point" != yes
+then
+   AC_MSG_NOTICE([Fixed point APIs disabled])
+   AC_DEFINE([PNG_NO_FIXED_POINT],[1],
+             [No support for fixed point APIs])
+fi
+if test "$enable_floating_arithmetic" != yes
+then
+   AC_MSG_NOTICE([Fixed point arithmetic will be used])
+   AC_DEFINE([PNG_NO_FLOATING_ARITHMETIC],[1],
+             [No support for floating point arithmetic])
+fi
 
 # The following is for pngvalid, to ensure it catches FP errors even on
 # platforms that don't enable FP exceptions, the function appears in the math
 # library (typically), it's not an error if it is not found.
-AC_CHECK_LIB([m], [feenableexcept])
-AC_CHECK_FUNCS([feenableexcept])
+AC_SEARCH_LIBS([feenableexcept],[m],AC_DEFINE([HAVE_FEENABLEEXCEPT],[1],
+      [Define to 1 if the feenableexcept function is available]))
+# Required by pngvalid:
+AC_SEARCH_LIBS([ceil],[m])
+
+# zlib checks and arguments;
+#  package zlib: argument is the name of the library
+#  zlib-header: the name of the header file
+#  zlib-prefix: the prefix for library functions, if any
+AC_ARG_WITH([zlib],
+   AS_HELP_STRING([[[--with-zlib=<name>]]],
+                  [base name of the zlib implementation (e.g. 'z')]),,
+   [with_zlib='z'])
+
+z_prefix=
+AC_ARG_WITH([zlib-prefix],
+   AS_HELP_STRING([[[--with-zlib-prefix=<prefix>]]],
+                  [prefix for zlib API functions]),
+   [if test "$withval" = "" -o "$withval" = yes; then
+      with_zlib_prefix='z_'
+      AC_DEFINE([PNG_ZLIB_PREFIX],[z_],[Standard zlib prefix])
+   elif test "${withval}" != "no"; then
+      AC_DEFINE_UNQUOTED([PNG_ZLIB_PREFIX],[${withval}],
+         [User defined zlib prefix])
+   else
+      with_zlib_prefix=
+   fi
+   if test "$with_zlib_prefix" != ""; then
+      AC_DEFINE([Z_PREFIX],[1],[Tell zlib.h to prefix function names])
+      z_prefix="#define Z_PREFIX 1"
+   fi])
+
+AC_ARG_WITH([zlib-header],
+   AS_HELP_STRING([[[--with-zlib-header='<file>' or '"file"']]],
+      [C include argument to locate zlib header file, e.g. <zlib.h>, take
+       care to put single quotes round this on the command line]),
+   [AC_DEFINE_UNQUOTED([PNG_ZLIB_HEADER],[${withval}],[Zlib header file])],
+   with_zlib_header='<zlib.h>')
+
+# AC_CHECK_LIB isn't quite enough for this because we need the special header
+# file too.
+LIBS="-l${with_zlib} $LIBS"
+AC_MSG_CHECKING(
+   [[Library '${with_zlib}' and #include ${with_zlib_header} for symbol '${with_zlib_prefix}zlibVersion']])
+AC_LINK_IFELSE(
+   [AC_LANG_PROGRAM(
+      [[
+${z_prefix}
+#include ${with_zlib_header}
+      ]],
+      [[(void)${with_zlib_prefix}zlibVersion();]])],
+   AC_MSG_RESULT([ok]),
+   AC_MSG_FAILURE([not found]))
 
 AC_MSG_CHECKING([if using Solaris linker])
 SLD=`$LD --version 2>&1 | grep Solaris`
@@ -210,7 +295,7 @@ AM_CONDITIONAL(HAVE_LD_VERSION_SCRIPT, test "$have_ld_version_script" = "yes")
 if test "$have_ld_version_script" = "yes"; then
     AC_MSG_CHECKING([for symbol prefix])
     SYMBOL_PREFIX=`echo "PREFIX=__USER_LABEL_PREFIX__" \
-                  | ${CPP-${CC-gcc} -E} - 2>&1 \
+                  | ${DFNCPP-${CC-gcc -E}} - 2>&1 \
                   | ${EGREP-grep} "^PREFIX=" \
                   | ${SED-sed} -e "s:^PREFIX=::" -e "s:__USER_LABEL_PREFIX__::"`
     AC_SUBST(SYMBOL_PREFIX)
@@ -297,188 +382,16 @@ AC_ARG_ENABLE([unversioned-libpng-config],
 AM_CONDITIONAL([DO_INSTALL_LIBPNG_CONFIG],
    [test "$enable_unversioned_libpng_config" != "no"])
 
-# HOST SPECIFIC OPTIONS
-# =====================
-#
-# ARM
-# ===
-#
-# ARM NEON (SIMD) support.
-
-AC_ARG_ENABLE([arm-neon],
-   AS_HELP_STRING([[[--enable-arm-neon]]],
-      [Enable ARM NEON optimizations: =no/off, check, api, yes/on:]
-      [no/off: disable the optimizations; check: use internal checking code]
-      [(deprecated and poorly supported); api: disable by default, enable by]
-      [a call to png_set_option; yes/on: turn on unconditionally.]
-      [If not specified: determined by the compiler.]),
-   [case "$enableval" in
-      no|off)
-         # disable the default enabling on __ARM_NEON__ systems:
-         AC_DEFINE([PNG_ARM_NEON_OPT], [0],
-                   [Disable ARM Neon optimizations])
-         # Prevent inclusion of the assembler files below:
-         enable_arm_neon=no;;
-      check)
-         AC_DEFINE([PNG_ARM_NEON_CHECK_SUPPORTED], [],
-                   [Check for ARM Neon support at run-time]);;
-      api)
-         AC_DEFINE([PNG_ARM_NEON_API_SUPPORTED], [],
-                   [Turn on ARM Neon optimizations at run-time]);;
-      yes|on)
-         AC_DEFINE([PNG_ARM_NEON_OPT], [2],
-                   [Enable ARM Neon optimizations])
-         AC_MSG_WARN([--enable-arm-neon: please specify 'check' or 'api', if]
-            [you want the optimizations unconditionally pass -mfpu=neon]
-            [to the compiler.]);;
-      *)
-         AC_MSG_ERROR([--enable-arm-neon=${enable_arm_neon}: invalid value])
-   esac])
-
-# Add ARM specific files to all builds where the host_cpu is arm ('arm*') or
-# where ARM optimizations were explicitly requested (this allows a fallback if a
-# future host CPU does not match 'arm*')
-
-AM_CONDITIONAL([PNG_ARM_NEON],
-   [test "$enable_arm_neon" != 'no' &&
-    case "$host_cpu" in
-      arm*|aarch64*) :;;
-      *)    test "$enable_arm_neon" != '';;
-    esac])
-
-# MIPS
-# ===
-#
-# MIPS MSA (SIMD) support.
-
-AC_ARG_ENABLE([mips-msa],
-   AS_HELP_STRING([[[--enable-mips-msa]]],
-      [Enable MIPS MSA optimizations: =no/off, check, api, yes/on:]
-      [no/off: disable the optimizations; check: use internal checking code]
-      [(deprecated and poorly supported); api: disable by default, enable by]
-      [a call to png_set_option; yes/on: turn on unconditionally.]
-      [If not specified: determined by the compiler.]),
-   [case "$enableval" in
-      no|off)
-         # disable the default enabling on __mips_msa systems:
-         AC_DEFINE([PNG_MIPS_MSA_OPT], [0],
-                   [Disable MIPS MSA optimizations])
-         # Prevent inclusion of the assembler files below:
-         enable_mips_msa=no;;
-      check)
-         AC_DEFINE([PNG_MIPS_MSA_CHECK_SUPPORTED], [],
-                   [Check for MIPS MSA support at run-time]);;
-      api)
-         AC_DEFINE([PNG_MIPS_MSA_API_SUPPORTED], [],
-                   [Turn on MIPS MSA optimizations at run-time]);;
-      yes|on)
-         AC_DEFINE([PNG_MIPS_MSA_OPT], [2],
-                   [Enable MIPS MSA optimizations])
-         AC_MSG_WARN([--enable-mips-msa: please specify 'check' or 'api', if]
-            [you want the optimizations unconditionally pass '-mmsa -mfp64']
-            [to the compiler.]);;
-      *)
-         AC_MSG_ERROR([--enable-mips-msa=${enable_mips_msa}: invalid value])
-   esac])
-
-# Add MIPS specific files to all builds where the host_cpu is mips ('mips*') or
-# where MIPS optimizations were explicitly requested (this allows a fallback if a
-# future host CPU does not match 'mips*')
-
-AM_CONDITIONAL([PNG_MIPS_MSA],
-   [test "$enable_mips_msa" != 'no' &&
-    case "$host_cpu" in
-      mipsel*|mips64el*) :;;
-    esac])
-
-# INTEL
-# =====
-#
-# INTEL SSE (SIMD) support.
-
-AC_ARG_ENABLE([intel-sse],
-   AS_HELP_STRING([[[--enable-intel-sse]]],
-      [Enable Intel SSE optimizations: =no/off, yes/on:]
-      [no/off: disable the optimizations;]
-      [yes/on: enable the optimizations.]
-      [If not specified: determined by the compiler.]),
-   [case "$enableval" in
-      no|off)
-         # disable the default enabling:
-         AC_DEFINE([PNG_INTEL_SSE_OPT], [0],
-                   [Disable Intel SSE optimizations])
-         # Prevent inclusion of the assembler files below:
-         enable_intel_sse=no;;
-      yes|on)
-         AC_DEFINE([PNG_INTEL_SSE_OPT], [1],
-                   [Enable Intel SSE optimizations]);;
-      *)
-         AC_MSG_ERROR([--enable-intel-sse=${enable_intel_sse}: invalid value])
-   esac])
-
-# Add Intel specific files to all builds where the host_cpu is Intel ('x86*')
-# or where Intel optimizations were explicitly requested (this allows a
-# fallback if a future host CPU does not match 'x86*')
-AM_CONDITIONAL([PNG_INTEL_SSE],
-   [test "$enable_intel_sse" != 'no' &&
-    case "$host_cpu" in
-      i?86|x86_64) :;;
-      *)    test "$enable_intel_sse" != '';;
-    esac])
-
-# PowerPC
-# ===
-#
-# PowerPC VSX (SIMD) support.
-
-AC_ARG_ENABLE([powerpc-vsx],
-AS_HELP_STRING([[[--enable-powerpc-vsx]]],
-      [Enable POWERPC VSX optimizations: =no/off, check, api, yes/on:]
-      [no/off: disable the optimizations; check: use internal checking code]
-      [api: disable by default, enable by a call to png_set_option]
-      [yes/on: turn on unconditionally.]
-      [If not specified: determined by the compiler.]),
-   [case "$enableval" in
-      no|off)
-         # disable the default enabling on __ppc64__ systems:
-         AC_DEFINE([PNG_POWERPC_VSX_OPT], [0],
-                   [Disable POWERPC VSX optimizations])
-         # Prevent inclusion of the platform specific files below:
-         enable_powerpc_vsx=no;;
-      check)
-         AC_DEFINE([PNG_POWERPC_VSX_CHECK_SUPPORTED], [],
-                   [Check for POWERPC VSX support at run-time])
-         AC_MSG_WARN([--enable-powerpc-vsx Please check contrib/powerpc/README file]
-            [for the list of supported OSes.]);;
-      api)
-         AC_DEFINE([PNG_POWERPC_VSX_API_SUPPORTED], [],
-                   [Turn on POWERPC VSX optimizations at run-time]);;
-      yes|on)
-         AC_DEFINE([PNG_POWERPC_VSX_OPT], [2],
-                   [Enable POWERPC VSX optimizations])
-         AC_MSG_WARN([--enable-powerpc-vsx: please specify 'check' or 'api', if]
-            [you want the optimizations unconditionally pass '-maltivec -mvsx']
-            [or '-mcpu=power8'to the compiler.]);;
-      *)
-         AC_MSG_ERROR([--enable-powerpc-vsx=${enable_powerpc_vsx}: invalid value])
-   esac])
-
-# Add PowerPC specific files to all builds where the host_cpu is powerpc('powerpc*') or
-# where POWERPC optimizations were explicitly requested (this allows a fallback if a
-# future host CPU does not match 'powerpc*')
-
-AM_CONDITIONAL([PNG_POWERPC_VSX],
-   [test "$enable_powerpc_vsx" != 'no' &&
-    case "$host_cpu" in
-      powerpc*|ppc64*) :;;
-    esac])
-
+# EXTENSIONS
+# Configure (and all other) support for extensions is included when autoconf is
+# run by virtue of the existence (or not) of the following file:
+sinclude([contrib/extensions.ac])
+AM_CONDITIONAL([LIBPNG_EXTENSIONS],[test -n "$libpng_automake_extensions"])
 
 AC_MSG_NOTICE([[Extra options for compiler: $PNG_COPTS]])
 
 # Config files, substituting as above
-AC_CONFIG_FILES([Makefile libpng.pc:libpng.pc.in])
-AC_CONFIG_FILES([libpng-config:libpng-config.in],
-   [chmod +x libpng-config])
+AC_CONFIG_FILES([Makefile libpng.pc])
+AC_CONFIG_FILES([libpng-config], [chmod +x libpng-config])
 
 AC_OUTPUT

contrib/arm-neon/android-ndk.c

@@ -17,7 +17,7 @@
  *
  * Documentation:
  *    http://www.kandroid.org/ndk/docs/CPU-ARM-NEON.html
- *    https://code.google.com/p/android/issues/detail?id=49065
+ *    http://code.google.com/p/android/issues/detail?id=49065
  *
  * NOTE: this requires that libpng is built against the Android NDK and linked
  * with an implementation of the Android ARM 'cpu-features' library.  The code

contrib/arm-neon/linux-auxv.c

@@ -7,7 +7,7 @@
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
  * and license in png.h
- *
+ * 
  * SEE contrib/arm-neon/README before reporting bugs
  *
  * STATUS: COMPILED, TESTED

contrib/arm-neon/linux.c

@@ -72,7 +72,7 @@ png_have_neon(png_structp png_ptr)
                      state = Colon;
                   break;
                }
-
+               
                /* did not match 'feature' */
                state = SkipLine;
                /* FALL THROUGH */

contrib/conftest/pngcp.dfa

@@ -1,57 +0,0 @@
-# pngcp.dfa
-#  Build time configuration of libpng
-#
-# Author: John Bowler
-# Copyright: (c) John Bowler, 2016
-# Usage rights:
-#  To the extent possible under law, the author has waived all copyright and
-#  related or neighboring rights to this work.  This work is published from:
-#  United States.
-#
-# Build libpng with support for pngcp.  This means just png_read_png,
-# png_write_png and small number of configuration settings.
-#
-everything = off
-
-# This option is specific to this configuration; it adds a #define to the
-# generated pnglibconf.h which turns on the (not portable) timing option for
-# pngcp.  Note that any option is automatically preceded by PNG_; there is no
-# way round this and this is deliberate.
-option PNGCP_TIMING
-
-# Because of the everything off above the option must also be turned on.  This
-# may not be done in one step because it is safer and avoids mis-spelled options
-# in user .dfa files to error out if an unrecognized option is turned on.
-option PNGCP_TIMING on
-
-# Options to turn on png_read_png and png_write_png:
-option INFO_IMAGE on
-option SEQUENTIAL_READ on
-option EASY_ACCESS on
-option WRITE on
-option WRITE_16BIT on
-option WRITE_FILTER on
-
-# pngcp needs this to preserve unknown chunks, switching all these on means that
-# pngcp can work without explicit known chunk reading suppport
-option UNKNOWN_CHUNKS on
-option SET_UNKNOWN_CHUNKS on
-option HANDLE_AS_UNKNOWN on
-option SAVE_UNKNOWN_CHUNKS on
-option WRITE_UNKNOWN_CHUNKS on
-
-# pngcp needs this to handle palette files with invalid indices:
-option CHECK_FOR_INVALID_INDEX on
-option GET_PALETTE_MAX on
-
-# Pre-libpng 1.7 pngcp has to stash text chunks manually, post 1.7 without this
-# text chunks should be handled as unknown ok.
-option TEXT on
-
-# this is used to turn off limits:
-option USER_LIMITS on
-option SET_USER_LIMITS on
-
-# these are are just required for specific customizations
-option WRITE_CUSTOMIZE_ZTXT_COMPRESSION on
-option WRITE_CUSTOMIZE_COMPRESSION on

contrib/examples/README.txt

@@ -21,4 +21,4 @@ ORIGINAL AUTHORS
     of the people below claim any rights with regard to the contents of this
     directory.
 
-    John Bowler <jbowler at acm.org>
+    John Bowler <jbowler@acm.org>

contrib/examples/pngtopng.c

@@ -1,12 +1,10 @@
 /*- pngtopng
  *
- * COPYRIGHT: Written by John Cunningham Bowler, 2011, 2017.
+ * COPYRIGHT: Written by John Cunningham Bowler, 2011.
  * To the extent possible under law, the author has waived all copyright and
  * related or neighboring rights to this work.  This work is published from:
  * United States.
  *
- * Last changed in libpng 1.6.29 [March 16, 2017]
- *
  * Read a PNG and write it out in a fixed format, using the 'simplified API'
  * that was introduced in libpng-1.6.0.
  *
@@ -61,27 +59,26 @@ int main(int argc, const char **argv)
                else
                   fprintf(stderr, "pngtopng: write %s: %s\n", argv[2],
                       image.message);
+
+               free(buffer);
             }
 
             else
+            {
                fprintf(stderr, "pngtopng: read %s: %s\n", argv[1],
                    image.message);
 
-            free(buffer);
+               /* This is the only place where a 'free' is required; libpng does
+                * the cleanup on error and success, but in this case we couldn't
+                * complete the read because of running out of memory.
+                */
+               png_image_free(&image);
+            }
          }
 
          else
-         {
             fprintf(stderr, "pngtopng: out of memory: %lu bytes\n",
                (unsigned long)PNG_IMAGE_SIZE(image));
-
-            /* This is the only place where a 'free' is required; libpng does
-             * the cleanup on error and success, but in this case we couldn't
-             * complete the read because of running out of memory and so libpng
-             * has not got to the point where it can do cleanup.
-             */
-            png_image_free(&image);
-         }
       }
 
       else

contrib/examples/simpleover.c

@@ -136,7 +136,7 @@ sprite_op(const struct sprite *sprite, int x_offset, int y_offset,
             /* This is the optimized Porter-Duff 'Over' operation, when the
              * input alpha is 0 the output is not changed.
              */
-            if (in_alpha > 0)
+            if (in_alpha > 0) 
             {
                png_uint_16 *out_pixel = sprite->buffer +
                   ((y+y_offset) * sprite->width + (x+x_offset))*4;
@@ -184,7 +184,7 @@ sprite_op(const struct sprite *sprite, int x_offset, int y_offset,
    }
 }
 
-static int
+static int 
 create_sprite(struct sprite *sprite, int *argc, const char ***argv)
 {
    /* Read the arguments and create this sprite. The sprite buffer has already
@@ -415,7 +415,7 @@ simpleover_process(png_imagep output, png_bytep out_buf, int argc,
             n = sscanf(argv[0], "--sprite=%u,%u,%" str(sprite_name_chars) "s%c",
                 &sprites[nsprites].width, &sprites[nsprites].height,
                 sprites[nsprites].name, &tombstone);
-
+            
             if ((n == 2 || n == 3) &&
                 sprites[nsprites].width > 0 && sprites[nsprites].height > 0)
             {
@@ -492,7 +492,7 @@ simpleover_process(png_imagep output, png_bytep out_buf, int argc,
                if (!add_sprite(output, out_buf, sprites+isprite, &argc, &argv))
                   goto out; /* error in add_sprite */
 
-               break;
+               break; 
             }
          }
 

contrib/extensions.ac

@@ -0,0 +1,16 @@
+# CONTRIBUTED EXTENSIONS
+#
+# The following extensions are part of a standard libpng distribution.  They are
+# not supported parts of libpng and they are not necessarily released under the
+# same license as libpng.  Please consult the relevant directory for copyright
+# and support information before uncommenting any of the lines below.
+#
+# After you change this file you MUST re-run "autoreconf" to regenerate the
+# configure script, otherwise the change may have no effect.
+#
+# Include the ARM NEON extension:
+#sinclude([contrib/neon/neon.ac])
+#
+# If any of the extensions require automake support (this is likely) automake
+# must include the file contrib/extensions.am
+#libpng_automake_extensions="yes"

contrib/extensions.am

@@ -0,0 +1,7 @@
+# CONTRIBUTED EXTENSIONS
+#
+# Please read the file "contrib/extensions.ac" before making changes to this
+# file.
+#
+# Include the ARM NEON extension:
+#include %reldir%/neon/neon.am

contrib/gregbook/Makefile.sgi

@@ -23,9 +23,9 @@
 
 # macros --------------------------------------------------------------------
 
-PNGINC = -I/usr/local/include/libpng16
-PNGLIB = -L/usr/local/lib -lpng16	  # dynamically linked against libpng
-#PNGLIB = /usr/local/lib/libpng16.a # statically linked against libpng
+PNGINC = -I/usr/local/include/libpng17
+PNGLIB = -L/usr/local/lib -lpng17	  # dynamically linked against libpng
+#PNGLIB = /usr/local/lib/libpng17.a # statically linked against libpng
 # or:
 #PNGINC = -I../..
 #PNGLIB = -L../.. -lpng

contrib/gregbook/Makefile.unx

@@ -26,14 +26,14 @@
 # macros --------------------------------------------------------------------
 
 #PNGDIR = /usr/local/lib
-#PNGINC = -I/usr/local/include/libpng16
-#PNGLIBd = -L$(PNGDIR) -lpng16 # dynamically linked, installed libpng
-#PNGLIBs = $(PNGDIR)/libpng16.a # statically linked, installed libpng
+#PNGINC = -I/usr/local/include/libpng17
+#PNGLIBd = -L$(PNGDIR) -lpng17 # dynamically linked, installed libpng
+#PNGLIBs = $(PNGDIR)/libpng17.a # statically linked, installed libpng
 # or:
 PNGDIR = ../..#	this one is for libpng-x.y.z/contrib/gregbook builds
 #PNGDIR = ../libpng
 PNGINC = -I$(PNGDIR)
-PNGLIBd = -Wl,-rpath,$(PNGDIR) -L$(PNGDIR) -lpng16	# dynamically linked
+PNGLIBd = -Wl,-rpath,$(PNGDIR) -L$(PNGDIR) -lpng17	# dynamically linked
 PNGLIBs = $(PNGDIR)/libpng.a		# statically linked, local libpng
 
 ZDIR = /usr/local/lib
@@ -59,16 +59,15 @@ INCS = $(PNGINC) $(ZINC) $(XINC)
 RLIBSd = $(PNGLIBd) $(ZLIBd) $(XLIB) -lm
 RLIBSs = $(PNGLIBs) $(ZLIBs) $(XLIB) -lm
 WLIBSd = $(PNGLIBd) $(ZLIBd) -lm
-WLIBSs = $(PNGLIBs) $(ZLIBs) -lm
+WLIBSs = $(PNGLIBs) $(ZLIBs)
 
 CC = gcc
 LD = gcc
 RM = rm -f
 CPPFLAGS = $(INCS) -DFEATURE_LOOP
 CFLAGS = -O -Wall
-#CFLAGS = -O -W -Wall -Wextra -pedantic -ansi
 # [note that -Wall is a gcc-specific compilation flag ("most warnings on")]
-# [-ansi, -pedantic, -Wextra, and -W can also be used]
+# [-ansi, -pedantic and -W can also be used]
 LDFLAGS =
 O = .o
 E =

contrib/gregbook/README

@@ -15,7 +15,7 @@ of PBMPLUS/NetPBM) and converts them to PNG.
 
 The source code for all three demo programs currently compiles under
 Unix, OpenVMS, and 32-bit Windows.  (Special thanks to Martin Zinser,
-zinser at decus.de, for making the necessary changes for OpenVMS and for
+zinser@decus.de, for making the necessary changes for OpenVMS and for
 providing an appropriate build script.)  Build instructions can be found
 below.
 
@@ -55,7 +55,7 @@ mation and links to the latest version of the source code, and Chapters
 13-15 of the book for detailed discussion of the three programs.
 
 Greg Roelofs
-https://pobox.com/~newt/greg_contact.html
+http://pobox.com/~newt/greg_contact.html
 16 March 2008
 
 

contrib/gregbook/readpng.c

@@ -100,8 +100,7 @@ int readpng_init(FILE *infile, ulg *pWidth, ulg *pHeight)
 
     /* could pass pointers to user-defined error handlers instead of NULLs: */
 
-    png_ptr = png_create_read_struct(png_get_libpng_ver(NULL), NULL, NULL,
-        NULL);
+    png_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING, NULL, NULL, NULL);
     if (!png_ptr)
         return 4;   /* out of memory */
 
@@ -264,12 +263,6 @@ uch *readpng_get_image(double display_exponent, int *pChannels, ulg *pRowbytes)
     *pRowbytes = rowbytes = png_get_rowbytes(png_ptr, info_ptr);
     *pChannels = (int)png_get_channels(png_ptr, info_ptr);
 
-    /* Guard against integer overflow */
-    if (height > ((size_t)(-1))/rowbytes) {
-        fprintf(stderr, "readpng:  image_data buffer would be too large\n",
-        return NULL;
-    }
-
     if ((image_data = (uch *)malloc(rowbytes*height)) == NULL) {
         png_destroy_read_struct(&png_ptr, &info_ptr, NULL);
         return NULL;

contrib/gregbook/readpng2.c

@@ -54,7 +54,7 @@
   ---------------------------------------------------------------------------
 
    Changelog:
-     2015-11-12 - Check return value of png_get_bKGD() (Glenn R-P)
+     %RDATE% - Check return value of png_get_bKGD() (Glenn R-P)
 
   ---------------------------------------------------------------------------*/
 

contrib/gregbook/readppm.c

@@ -154,18 +154,13 @@ uch *readpng_get_image(double display_exponent, int *pChannels, ulg *pRowbytes)
     *pRowbytes = rowbytes = channels*width;
     *pChannels = channels;
 
-    Trace((stderr, "readpng_get_image:  rowbytes = %ld, height = %ld\n", rowbytes, height));
-
-    /* Guard against integer overflow */
-    if (height > ((size_t)(-1))/rowbytes) {
-        fprintf(stderr, PROGNAME ":  image_data buffer would be too large\n",
-        return NULL;
-    }
-
     if ((image_data = (uch *)malloc(rowbytes*height)) == NULL) {
         return NULL;
     }
 
+    Trace((stderr, "readpng_get_image:  rowbytes = %ld, height = %ld\n", rowbytes, height));
+
+
     /* now we can go ahead and just read the whole image */
 
     if (fread(image_data, 1L, rowbytes*height, saved_infile) <

contrib/gregbook/rpng-win.c

@@ -182,7 +182,7 @@ int WINAPI WinMain(HINSTANCE hInst, HINSTANCE hPrevInst, PSTR cmd, int showmode)
 #ifndef __CYGWIN__
     /* First reenable console output, which normally goes to the bit bucket
      * for windowed apps.  Closing the console window will terminate the
-     * app.  Thanks to David.Geldreich at realviz.com for supplying the magical
+     * app.  Thanks to David.Geldreich@realviz.com for supplying the magical
      * incantation. */
 
     AllocConsole();
@@ -496,12 +496,6 @@ static int rpng_win_create_window(HINSTANCE hInst, int showmode)
 
     wimage_rowbytes = ((3*image_width + 3L) >> 2) << 2;
 
-    /* Guard against integer overflow */
-    if (image_height > ((size_t)(-1))/wimage_rowbytes) {
-        fprintf(stderr, PROGNAME ":  image_data buffer would be too large\n",
-        return 4;   /* fail */
-    }
-
     if (!(dib = (uch *)malloc(sizeof(BITMAPINFOHEADER) +
                               wimage_rowbytes*image_height)))
     {

contrib/gregbook/rpng-x.c

@@ -26,8 +26,6 @@
     - 1.14:  added support for X resources (thanks to Gerhard Niklasch)
     - 2.00:  dual-licensed (added GNU GPL)
     - 2.01:  fixed improper display of usage screen on PNG error(s)
-    - 2.02:  Added "void(argc);" statement to quiet pedantic compiler warnings
-             about unused variable (GR-P)
 
   ---------------------------------------------------------------------------
 
@@ -82,7 +80,7 @@
 
 #define PROGNAME  "rpng-x"
 #define LONGNAME  "Simple PNG Viewer for X"
-#define VERSION   "2.02 of 15 June 2014"
+#define VERSION   "2.01 of 16 March 2008"
 #define RESNAME   "rpng"        /* our X resource application name */
 #define RESCLASS  "Rpng"        /* our X resource class name */
 
@@ -281,17 +279,15 @@ int main(int argc, char **argv)
           "Usage:  %s [-display xdpy] [-gamma exp] [-bgcolor bg] file.png\n"
           "    xdpy\tname of the target X display (e.g., ``hostname:0'')\n"
           "    exp \ttransfer-function exponent (``gamma'') of the display\n"
-          "\t\t  system in floating-point format (e.g., ``%.1f''); equal\n",
-          PROGNAME, default_display_exponent);
-
-        fprintf(stderr, "\n"
+          "\t\t  system in floating-point format (e.g., ``%.1f''); equal\n"
           "\t\t  to the product of the lookup-table exponent (varies)\n"
           "\t\t  and the CRT exponent (usually 2.2); must be positive\n"
           "    bg  \tdesired background color in 7-character hex RGB format\n"
           "\t\t  (e.g., ``#ff7700'' for orange:  same as HTML colors);\n"
           "\t\t  used with transparent images\n"
           "\nPress Q, Esc or mouse button 1 (within image window, after image\n"
-          "is displayed) to quit.\n");
+          "is displayed) to quit.\n"
+          "\n", PROGNAME, default_display_exponent);
         exit(1);
     }
 
@@ -423,8 +419,6 @@ int main(int argc, char **argv)
 
     rpng_x_cleanup();
 
-    (void)argc; /* Unused */
-
     return 0;
 }
 

contrib/gregbook/rpng2-win.c

@@ -33,8 +33,6 @@
     - 2.02:  fixed improper display of usage screen on PNG error(s); fixed
               unexpected-EOF and file-read-error cases
     - 2.03:  removed runtime MMX-enabling/disabling and obsolete -mmx* options
-    - 2.04:
-             (GR-P)
 
   ---------------------------------------------------------------------------
 
@@ -301,7 +299,7 @@ int WINAPI WinMain(HINSTANCE hInst, HINSTANCE hPrevInst, PSTR cmd, int showmode)
 #ifndef __CYGWIN__
     /* Next reenable console output, which normally goes to the bit bucket
      * for windowed apps.  Closing the console window will terminate the
-     * app.  Thanks to David.Geldreich at realviz.com for supplying the magical
+     * app.  Thanks to David.Geldreich@realviz.com for supplying the magical
      * incantation. */
 
     AllocConsole();
@@ -650,13 +648,6 @@ static void rpng2_win_init()
     Trace((stderr, "  width  = %ld\n", rpng2_info.width))
     Trace((stderr, "  height = %ld\n", rpng2_info.height))
 
-    /* Guard against integer overflow */
-    if (rpng2_info.height > ((size_t)(-1))/rowbytes) {
-        fprintf(stderr, PROGNAME ":  image_data buffer would be too large\n",
-        readpng2_cleanup(&rpng2_info);
-        return;
-    }
-
     rpng2_info.image_data = (uch *)malloc(rowbytes * rpng2_info.height);
     if (!rpng2_info.image_data) {
         readpng2_cleanup(&rpng2_info);

contrib/gregbook/rpng2-x.c

@@ -496,12 +496,12 @@ int main(int argc, char **argv)
           "\t\t  transparent images; overrides -bgcolor\n",
           num_bgpat-1);
 #ifdef FEATURE_LOOP
-        fprintf(stderr,
+        fprintf(stderr, 
           "    -loop\tloops through background images after initial display\n"
           "\t\t  is complete (depends on -bgpat)\n"
           "    sec \tseconds to display each background image (default = 2)\n");
 #endif
-        fprintf(stderr,
+        fprintf(stderr, 
           "    dur \tduration in microseconds to wait after displaying each\n"
           "\t\t  row (for demo purposes)\n"
           "    -timing\tenables delay for every block read, to simulate modem\n"
@@ -780,13 +780,6 @@ static void rpng2_x_init(void)
     Trace((stderr, "  width  = %ld\n", rpng2_info.width))
     Trace((stderr, "  height = %ld\n", rpng2_info.height))
 
-    /* Guard against integer overflow */
-    if (rpng2_info.height > ((size_t)(-1))/rpng2_info.rowbytes) {
-        fprintf(stderr, PROGNAME ":  image_data buffer would be too large\n");
-        readpng2_cleanup(&rpng2_info);
-        return;
-    }
-
     rpng2_info.image_data = (uch *)malloc(rowbytes * rpng2_info.height);
     if (!rpng2_info.image_data) {
         readpng2_cleanup(&rpng2_info);

contrib/gregbook/wpng.c

@@ -702,17 +702,7 @@ int main(int argc, char **argv)
     if (wpng_info.interlaced) {
         long i;
         ulg bytes;
-        ulg image_bytes;
-
-        /* Guard against integer overflow */
-        if (wpng_info_height > ((size_t)(-1)/rowbytes) {
-            fprintf(stderr, PROGNAME ":  image_data buffer too large\n");
-            writepng_cleanup(&wpng_info);
-            wpng_cleanup();
-            exit(5);
-        }
-
-        image_bytes = rowbytes * wpng_info.height;   /* overflow? */
+        ulg image_bytes = rowbytes * wpng_info.height;   /* overflow? */
 
         wpng_info.image_data = (uch *)malloc(image_bytes);
         wpng_info.row_pointers = (uch **)malloc(wpng_info.height*sizeof(uch *));

contrib/gregbook/writepng.c

@@ -90,7 +90,7 @@ int writepng_init(mainprog_info *mainprog_ptr)
 
     /* could also replace libpng warning-handler (final NULL), but no need: */
 
-    png_ptr = png_create_write_struct(png_get_libpng_ver(NULL), mainprog_ptr,
+    png_ptr = png_create_write_struct(PNG_LIBPNG_VER_STRING, mainprog_ptr,
       writepng_error_handler, NULL);
     if (!png_ptr)
         return 4;   /* out of memory */

contrib/libtests/makepng.c

@@ -3,7 +3,7 @@
 /* Copyright: */
 #define COPYRIGHT "\251 2013,2015 John Cunningham Bowler"
 /*
- * Last changed in libpng 1.6.20 [November 24, 2015]
+ * Last changed in libpng 1.7.0 [(PENDING RELEASE)]
  *
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
@@ -392,7 +392,7 @@ generate_row(png_bytep row, size_t rowbytes, unsigned int y, int color_type,
    unsigned int *colors, int small)
 {
    int filters = 0; /* file *MASK*, 0 means the default, not NONE */
-   png_uint_32 size_max =
+   png_uint_32 size_max = 
       image_size_of_type(color_type, bit_depth, colors, small)-1;
    png_uint_32 depth_max = (1U << bit_depth)-1; /* up to 65536 */
 
@@ -526,7 +526,7 @@ generate_row(png_bytep row, size_t rowbytes, unsigned int y, int color_type,
             assert(0/*NOT REACHED*/);
       }
    }
-
+      
    else switch (channels_of_type(color_type))
    {
    /* 1 channel: a square image with a diamond, the least luminous colors are on
@@ -1913,7 +1913,7 @@ main(int argc, char **argv)
       };
 
       chunk_insert *new_insert;
-
+      
       new_insert = add_tEXt("Copyright", copyright);
       if (new_insert != NULL)
       {

contrib/libtests/pngimage.c

@@ -1,8 +1,8 @@
 /* pngimage.c
  *
- * Copyright (c) 2015,2016 John Cunningham Bowler
+ * Copyright (c) 2015 John Cunningham Bowler
  *
- * Last changed in libpng 1.6.24 [August 4, 2016]
+ * Last changed in libpng 1.6.18 [(PENDING RELEASE)]
  *
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
@@ -36,28 +36,7 @@
 #  include <setjmp.h> /* because png.h did *not* include this */
 #endif
 
-/* 1.6.1 added support for the configure test harness, which uses 77 to indicate
- * a skipped test, in earlier versions we need to succeed on a skipped test, so:
- */
-#if PNG_LIBPNG_VER >= 10601 && defined(HAVE_CONFIG_H)
-#  define SKIP 77
-#else
-#  define SKIP 0
-#endif
-
-#if PNG_LIBPNG_VER < 10700
-   /* READ_PNG and WRITE_PNG were not defined, so: */
-#  ifdef PNG_INFO_IMAGE_SUPPORTED
-#     ifdef PNG_SEQUENTIAL_READ_SUPPORTED
-#        define PNG_READ_PNG_SUPPORTED
-#     endif /* SEQUENTIAL_READ */
-#     ifdef PNG_WRITE_SUPPORTED
-#        define PNG_WRITE_PNG_SUPPORTED
-#     endif /* WRITE */
-#  endif /* INFO_IMAGE */
-#endif /* pre 1.7.0 */
-
-#ifdef PNG_READ_PNG_SUPPORTED
+#if defined(PNG_INFO_IMAGE_SUPPORTED) && defined(PNG_SEQUENTIAL_READ_SUPPORTED)
 /* If a transform is valid on both read and write this implies that if the
  * transform is applied to read it must also be applied on write to produce
  * meaningful data.  This is because these transforms when performed on read
@@ -406,7 +385,7 @@ buffer_destroy(struct buffer *buffer)
    buffer_destroy_list(list);
 }
 
-#ifdef PNG_WRITE_PNG_SUPPORTED
+#ifdef PNG_WRITE_SUPPORTED
 static void
 buffer_start_write(struct buffer *buffer)
 {
@@ -576,7 +555,7 @@ struct display
    png_structp    read_pp;
    png_infop      read_ip;
 
-#  ifdef PNG_WRITE_PNG_SUPPORTED
+#  ifdef PNG_WRITE_SUPPORTED
       /* Used to write a new image (the original info_ptr is used) */
       png_structp   write_pp;
       struct buffer written_file;   /* where the file gets written */
@@ -603,7 +582,7 @@ display_init(struct display *dp)
    dp->read_ip = NULL;
    buffer_init(&dp->original_file);
 
-#  ifdef PNG_WRITE_PNG_SUPPORTED
+#  ifdef PNG_WRITE_SUPPORTED
       dp->write_pp = NULL;
       buffer_init(&dp->written_file);
 #  endif
@@ -616,7 +595,7 @@ display_clean_read(struct display *dp)
       png_destroy_read_struct(&dp->read_pp, &dp->read_ip, NULL);
 }
 
-#ifdef PNG_WRITE_PNG_SUPPORTED
+#ifdef PNG_WRITE_SUPPORTED
 static void
 display_clean_write(struct display *dp)
 {
@@ -628,7 +607,7 @@ display_clean_write(struct display *dp)
 static void
 display_clean(struct display *dp)
 {
-#  ifdef PNG_WRITE_PNG_SUPPORTED
+#  ifdef PNG_WRITE_SUPPORTED
       display_clean_write(dp);
 #  endif
    display_clean_read(dp);
@@ -646,7 +625,7 @@ static void
 display_destroy(struct display *dp)
 {
     /* Release any memory held in the display. */
-#  ifdef PNG_WRITE_PNG_SUPPORTED
+#  ifdef PNG_WRITE_SUPPORTED
       buffer_destroy(&dp->written_file);
 #  endif
 
@@ -968,7 +947,7 @@ update_display(struct display *dp)
 
          if ((transform_info[i].valid_chunks == 0 ||
                (transform_info[i].valid_chunks & chunks) != 0) &&
-            (transform_info[i].color_mask_required & ct) ==
+            (transform_info[i].color_mask_required & ct) == 
                transform_info[i].color_mask_required &&
             (transform_info[i].color_mask_absent & ct) == 0 &&
             (transform_info[i].bit_depths & bd) != 0 &&
@@ -1026,7 +1005,7 @@ compare_read(struct display *dp, int applied_transforms)
    {
       unsigned long chunks =
          png_get_valid(dp->read_pp, dp->read_ip, 0xffffffff);
-
+      
       if (chunks != dp->chunks)
          display_log(dp, APP_FAIL, "PNG chunks changed from 0x%lx to 0x%lx",
             (unsigned long)dp->chunks, chunks);
@@ -1093,7 +1072,6 @@ compare_read(struct display *dp, int applied_transforms)
       }
 
       else
-#     ifdef PNG_sBIT_SUPPORTED
       {
          unsigned long y;
          int bpp;   /* bits-per-pixel then bytes-per-pixel */
@@ -1255,16 +1233,12 @@ compare_read(struct display *dp, int applied_transforms)
             }
          } /* for y */
       }
-#     else /* !sBIT */
-         display_log(dp, INTERNAL_ERROR,
-               "active shift transform but no sBIT support");
-#     endif /* !sBIT */
    }
 
    return 1; /* compare succeeded */
 }
 
-#ifdef PNG_WRITE_PNG_SUPPORTED
+#ifdef PNG_WRITE_SUPPORTED
 static void
 buffer_write(struct display *dp, struct buffer *buffer, png_bytep data,
    png_size_t size)
@@ -1363,7 +1337,7 @@ write_png(struct display *dp, png_infop ip, int transforms)
     */
    display_clean_write(dp);
 }
-#endif /* WRITE_PNG */
+#endif /* WRITE_SUPPORTED */
 
 static int
 skip_transform(struct display *dp, int tr)
@@ -1425,7 +1399,7 @@ test_one_file(struct display *dp, const char *filename)
          return; /* no point testing more */
    }
 
-#ifdef PNG_WRITE_PNG_SUPPORTED
+#ifdef PNG_WRITE_SUPPORTED
    /* Second test: write the original PNG data out to a new file (to test the
     * write side) then read the result back in and make sure that it hasn't
     * changed.
@@ -1466,7 +1440,7 @@ test_one_file(struct display *dp, const char *filename)
           * out and read it back in again (without the reversible transforms)
           * we should get back to the place where we started.
           */
-#ifdef PNG_WRITE_PNG_SUPPORTED
+#ifdef PNG_WRITE_SUPPORTED
          if ((current & write_transforms) == current)
          {
             /* All transforms reversible: write the PNG with the transformations
@@ -1702,11 +1676,11 @@ main(const int argc, const char * const * const argv)
       return errors != 0;
    }
 }
-#else /* !READ_PNG */
+#else /* !PNG_INFO_IMAGE_SUPPORTED || !PNG_READ_SUPPORTED */
 int
 main(void)
 {
    fprintf(stderr, "pngimage: no support for png_read/write_image\n");
-   return SKIP;
+   return 77;
 }
 #endif

contrib/libtests/pngstest-errors.h

@@ -1,6 +1,6 @@
 /* contrib/libtests/pngstest-errors.h
  *
- * BUILT USING: libpng version 1.6.19beta03 - September 25, 2015
+ * BUILT USING: libpng version 1.7.0beta66 - September 18, 2015
  *
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
@@ -22,7 +22,7 @@ static png_uint_16 gpc_error[16/*in*/][16/*out*/][4/*a*/] =
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 },
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }
  }, { /* input: sRGB-gray+alpha */
-  { 0, 19, 0, 0 }, { 0, 0, 0, 0 }, { 0, 20, 0, 0 }, { 0, 0, 0, 0 },
+  { 0, 19, 1, 0 }, { 0, 0, 0, 0 }, { 0, 20, 1, 0 }, { 0, 0, 0, 0 },
   { 0, 897, 788, 0 }, { 0, 897, 788, 0 }, { 0, 897, 788, 0 }, { 0, 897, 788, 0 },
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 },
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }
@@ -32,8 +32,8 @@ static png_uint_16 gpc_error[16/*in*/][16/*out*/][4/*a*/] =
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 },
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }
  }, { /* input: sRGB-rgb+alpha */
-  { 0, 16, 17, 0 }, { 0, 17, 17, 0 }, { 0, 19, 0, 0 }, { 0, 0, 0, 0 },
-  { 0, 850, 875, 0 }, { 0, 850, 875, 0 }, { 0, 897, 788, 0 }, { 0, 897, 788, 0 },
+  { 0, 14, 13, 0 }, { 0, 15, 13, 0 }, { 0, 19, 1, 0 }, { 0, 0, 0, 0 },
+  { 0, 832, 764, 0 }, { 0, 832, 764, 0 }, { 0, 897, 788, 0 }, { 0, 897, 788, 0 },
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 },
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }
  }, { /* input: linear-gray */
@@ -67,13 +67,13 @@ static png_uint_16 gpc_error[16/*in*/][16/*out*/][4/*a*/] =
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 },
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }
  }, { /* input: color-mapped-sRGB-rgb */
-  { 0, 0, 13, 0 }, { 0, 0, 13, 0 }, { 0, 0, 9, 0 }, { 0, 0, 9, 0 },
+  { 0, 0, 13, 0 }, { 0, 0, 13, 0 }, { 0, 0, 8, 0 }, { 0, 0, 8, 0 },
   { 0, 0, 673, 0 }, { 0, 0, 673, 0 }, { 0, 0, 674, 0 }, { 0, 0, 674, 0 },
   { 0, 0, 1, 0 }, { 0, 0, 1, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 },
   { 0, 0, 460, 0 }, { 0, 0, 460, 0 }, { 0, 0, 263, 0 }, { 0, 0, 263, 0 }
  }, { /* input: color-mapped-sRGB-rgb+alpha */
-  { 0, 6, 8, 0 }, { 0, 7, 8, 0 }, { 0, 75, 9, 0 }, { 0, 9, 9, 0 },
-  { 0, 585, 427, 0 }, { 0, 585, 427, 0 }, { 0, 717, 514, 0 }, { 0, 717, 514, 0 },
+  { 0, 6, 8, 0 }, { 0, 7, 8, 0 }, { 0, 75, 8, 0 }, { 0, 9, 8, 0 },
+  { 0, 585, 427, 0 }, { 0, 585, 427, 0 }, { 0, 717, 584, 0 }, { 0, 717, 584, 0 },
   { 0, 1, 1, 0 }, { 0, 1, 1, 0 }, { 0, 1, 0, 0 }, { 0, 0, 0, 0 },
   { 0, 13323, 460, 0 }, { 0, 427, 460, 0 }, { 0, 16480, 263, 0 }, { 0, 243, 263, 0 }
  }, { /* input: color-mapped-linear-gray */
@@ -107,7 +107,7 @@ static png_uint_16 gpc_error_via_linear[16][4/*out*/][4] =
  }, { /* input: sRGB-rgb */
   { 0, 0, 20, 0 }, { 0, 0, 20, 0 }, { 0, 0, 15, 0 }, { 0, 0, 15, 0 }
  }, { /* input: sRGB-rgb+alpha */
-  { 0, 16, 17, 0 }, { 0, 187, 17, 0 }, { 0, 15, 15, 0 }, { 0, 186, 15, 0 }
+  { 0, 12, 14, 0 }, { 0, 180, 14, 0 }, { 0, 15, 15, 0 }, { 0, 186, 15, 0 }
  }, { /* input: linear-gray */
   { 0, 0, 1, 0 }, { 0, 0, 1, 0 }, { 0, 0, 1, 0 }, { 0, 0, 1, 0 }
  }, { /* input: linear-gray+alpha */
@@ -123,7 +123,7 @@ static png_uint_16 gpc_error_via_linear[16][4/*out*/][4] =
  }, { /* input: color-mapped-sRGB-rgb */
   { 0, 0, 13, 0 }, { 0, 0, 13, 0 }, { 0, 0, 14, 0 }, { 0, 0, 14, 0 }
  }, { /* input: color-mapped-sRGB-rgb+alpha */
-  { 0, 4, 8, 0 }, { 0, 9, 8, 0 }, { 0, 9, 5, 0 }, { 0, 32, 5, 0 }
+  { 0, 4, 8, 0 }, { 0, 9, 8, 0 }, { 0, 9, 9, 0 }, { 0, 32, 9, 0 }
  }, { /* input: color-mapped-linear-gray */
   { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }, { 0, 0, 0, 0 }
  }, { /* input: color-mapped-linear-gray+alpha */
@@ -138,27 +138,27 @@ static png_uint_16 gpc_error_to_colormap[8/*i*/][8/*o*/][4] =
 {
  { /* input: sRGB-gray */
   { 0, 0, 9, 0 }, { 0, 0, 9, 0 }, { 0, 0, 9, 0 }, { 0, 0, 9, 0 },
-  { 0, 0, 560, 0 }, { 0, 0, 560, 0 }, { 0, 0, 560, 0 }, { 0, 0, 560, 0 }
+  { 0, 0, 742, 0 }, { 0, 0, 742, 0 }, { 0, 0, 742, 0 }, { 0, 0, 742, 0 }
  }, { /* input: sRGB-gray+alpha */
-  { 0, 19, 9, 0 }, { 0, 255, 9, 25 }, { 0, 88, 9, 0 }, { 0, 255, 9, 25 },
-  { 0, 1012, 928, 0 }, { 0, 16026, 928, 6425 }, { 0, 1012, 928, 0 }, { 0, 16026, 928, 6425 }
+  { 0, 19, 6, 0 }, { 0, 255, 6, 25 }, { 0, 88, 6, 0 }, { 0, 255, 6, 25 },
+  { 0, 1012, 911, 0 }, { 0, 16026, 911, 6425 }, { 0, 1012, 911, 0 }, { 0, 16026, 911, 6425 }
  }, { /* input: sRGB-rgb */
   { 0, 0, 19, 0 }, { 0, 0, 19, 0 }, { 0, 0, 25, 0 }, { 0, 0, 25, 0 },
   { 0, 0, 962, 0 }, { 0, 0, 962, 0 }, { 0, 0, 13677, 0 }, { 0, 0, 13677, 0 }
  }, { /* input: sRGB-rgb+alpha */
   { 0, 63, 77, 0 }, { 0, 255, 19, 25 }, { 0, 225, 25, 0 }, { 0, 255, 25, 67 },
-  { 0, 17534, 18491, 0 }, { 0, 15736, 2824, 6425 }, { 0, 14019, 13677, 0 }, { 0, 50115, 13677, 17219 }
+  { 0, 17534, 18491, 0 }, { 0, 15998, 2824, 6425 }, { 0, 14103, 13677, 0 }, { 0, 50115, 13677, 17219 }
  }, { /* input: linear-gray */
   { 0, 0, 73, 0 }, { 0, 0, 73, 0 }, { 0, 0, 73, 0 }, { 0, 0, 73, 0 },
   { 0, 0, 18817, 0 }, { 0, 0, 18817, 0 }, { 0, 0, 18817, 0 }, { 0, 0, 18817, 0 }
  }, { /* input: linear-gray+alpha */
-  { 0, 74, 74, 0 }, { 0, 255, 74, 25 }, { 0, 99, 74, 0 }, { 0, 255, 74, 25 },
+  { 0, 74, 74, 0 }, { 0, 255, 74, 25 }, { 0, 100, 74, 0 }, { 0, 255, 74, 25 },
   { 0, 18919, 18907, 0 }, { 0, 24549, 18907, 6553 }, { 0, 18919, 18907, 0 }, { 0, 24549, 18907, 6553 }
  }, { /* input: linear-rgb */
   { 0, 0, 73, 0 }, { 0, 0, 73, 0 }, { 0, 0, 98, 0 }, { 0, 0, 98, 0 },
   { 0, 0, 18664, 0 }, { 0, 0, 18664, 0 }, { 0, 0, 24998, 0 }, { 0, 0, 24998, 0 }
  }, { /* input: linear-rgb+alpha */
-  { 0, 181, 196, 0 }, { 0, 255, 61, 25 }, { 206, 187, 98, 0 }, { 0, 255, 98, 67 },
+  { 0, 181, 196, 0 }, { 0, 255, 61, 25 }, { 206, 192, 98, 0 }, { 0, 255, 98, 67 },
   { 0, 18141, 18137, 0 }, { 0, 17494, 17504, 6553 }, { 0, 24979, 24992, 0 }, { 0, 49172, 24992, 17347 }
  }
 };

contrib/libtests/pngstest.c

@@ -1,9 +1,9 @@
 /*-
  * pngstest.c
  *
- * Copyright (c) 2013-2016 John Cunningham Bowler
+ * Copyright (c) 2013-2015 John Cunningham Bowler
  *
- * Last changed in libpng 1.6.24 [August 4, 2016]
+ * Last changed in libpng 1.6.19 [(PENDING RELEASE)]
  *
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
@@ -35,15 +35,6 @@
 #  include "../../png.h"
 #endif
 
-/* 1.6.1 added support for the configure test harness, which uses 77 to indicate
- * a skipped test, in earlier versions we need to succeed on a skipped test, so:
- */
-#if PNG_LIBPNG_VER >= 10601 && defined(HAVE_CONFIG_H)
-#  define SKIP 77
-#else
-#  define SKIP 0
-#endif
-
 #ifdef PNG_SIMPLIFIED_READ_SUPPORTED /* Else nothing can be done */
 #include "../tools/sRGB.h"
 
@@ -108,18 +99,10 @@ make_random_bytes(png_uint_32* seed, void* pv, size_t size)
    seed[1] = u1;
 }
 
-static png_uint_32 color_seed[2];
-
-static void
-reseed(void)
-{
-   color_seed[0] = 0x12345678U;
-   color_seed[1] = 0x9abcdefU;
-}
-
 static void
 random_color(png_colorp color)
 {
+   static png_uint_32 color_seed[2] = { 0x12345678, 0x9abcdef };
    make_random_bytes(color_seed, color, sizeof *color);
 }
 
@@ -324,7 +307,7 @@ compare_16bit(int v1, int v2, int error_limit, int multiple_algorithms)
 }
 #endif /* unused */
 
-#define USE_FILE 1       /* else memory */
+#define READ_FILE 1      /* else memory */
 #define USE_STDIO 2      /* else use file name */
 #define STRICT 4         /* fail on warnings too */
 #define VERBOSE 8
@@ -333,19 +316,16 @@ compare_16bit(int v1, int v2, int error_limit, int multiple_algorithms)
 #define ACCUMULATE 64
 #define FAST_WRITE 128
 #define sRGB_16BIT 256
-#define NO_RESEED  512   /* do not reseed on each new file */
-#define GBG_ERROR 1024   /* do not ignore the gamma+background_rgb_to_gray
-                          * libpng warning. */
 
 static void
 print_opts(png_uint_32 opts)
 {
-   if (opts & USE_FILE)
+   if (opts & READ_FILE)
       printf(" --file");
    if (opts & USE_STDIO)
       printf(" --stdio");
-   if (!(opts & STRICT))
-      printf(" --nostrict");
+   if (opts & STRICT)
+      printf(" --strict");
    if (opts & VERBOSE)
       printf(" --verbose");
    if (opts & KEEP_TMPFILES)
@@ -358,12 +338,6 @@ print_opts(png_uint_32 opts)
       printf(" --slow");
    if (opts & sRGB_16BIT)
       printf(" --sRGB-16bit");
-   if (opts & NO_RESEED)
-      printf(" --noreseed");
-#if PNG_LIBPNG_VER < 10700 /* else on by default */
-   if (opts & GBG_ERROR)
-      printf(" --fault-gbg-warning");
-#endif
 }
 
 #define FORMAT_NO_CHANGE 0x80000000 /* additional flag */
@@ -767,15 +741,8 @@ checkopaque(Image *image)
       return logerror(image, image->file_name, ": opaque not NULL", "");
    }
 
-   /* Separate out the gamma+background_rgb_to_gray warning because it may
-    * produce opaque component errors:
-    */
-   else if (image->image.warning_or_error != 0 &&
-            (strcmp(image->image.message,
-               "libpng does not support gamma+background+rgb_to_gray") == 0 ?
-                  (image->opts & GBG_ERROR) != 0 : (image->opts & STRICT) != 0))
-      return logerror(image, image->file_name, (image->opts & GBG_ERROR) != 0 ?
-                      " --fault-gbg-warning" : " --strict", "");
+   else if (image->image.warning_or_error != 0 && (image->opts & STRICT) != 0)
+      return logerror(image, image->file_name, " --strict", "");
 
    else
       return 1;
@@ -3041,14 +3008,14 @@ read_file(Image *image, png_uint_32 format, png_const_colorp background)
 static int
 read_one_file(Image *image)
 {
-   if (!(image->opts & USE_FILE) || (image->opts & USE_STDIO))
+   if (!(image->opts & READ_FILE) || (image->opts & USE_STDIO))
    {
       /* memory or stdio. */
       FILE *f = fopen(image->file_name, "rb");
 
       if (f != NULL)
       {
-         if (image->opts & USE_FILE)
+         if (image->opts & READ_FILE)
             image->input_file = f;
 
          else /* memory */
@@ -3129,8 +3096,7 @@ write_one_file(Image *output, Image *image, int convert_to_8bit)
 
    if (image->opts & USE_STDIO)
    {
-#ifdef PNG_SIMPLIFIED_WRITE_STDIO_SUPPORTED
-#ifndef __COVERITY__
+#ifndef PNG_USE_MKSTEMP
       FILE *f = tmpfile();
 #else
       /* Experimental. Coverity says tmpfile() is insecure because it
@@ -3164,7 +3130,7 @@ write_one_file(Image *output, Image *image, int convert_to_8bit)
         (void) unlink(tmpfile);
       }
 #endif
-
+      
       if (f != NULL)
       {
          if (png_image_write_to_stdio(&image->image, f, convert_to_8bit,
@@ -3192,14 +3158,10 @@ write_one_file(Image *output, Image *image, int convert_to_8bit)
 
       else
          return logerror(image, "tmpfile", ": open: ", strerror(errno));
-#else /* SIMPLIFIED_WRITE_STDIO */
-      return logerror(image, "tmpfile", ": open: unsupported", "");
-#endif /* SIMPLIFIED_WRITE_STDIO */
    }
 
-   else if (image->opts & USE_FILE)
+   else
    {
-#ifdef PNG_SIMPLIFIED_WRITE_STDIO_SUPPORTED
       static int counter = 0;
       char name[32];
 
@@ -3219,51 +3181,6 @@ write_one_file(Image *output, Image *image, int convert_to_8bit)
 
       else
          return logerror(image, name, ": write failed", "");
-#else /* SIMPLIFIED_WRITE_STDIO */
-      return logerror(image, "stdio", ": open: unsupported", "");
-#endif /* SIMPLIFIED_WRITE_STDIO */
-   }
-
-   else /* use memory */
-   {
-      png_alloc_size_t size;
-
-      if (png_image_write_get_memory_size(image->image, size, convert_to_8bit,
-               image->buffer+16, (png_int_32)image->stride, image->colormap))
-      {
-         /* This is non-fatal but ignoring it was causing serious problems in
-          * the macro to be ignored:
-          */
-         if (size > PNG_IMAGE_PNG_SIZE_MAX(image->image))
-            return logerror(image, "memory", ": PNG_IMAGE_SIZE_MAX wrong", "");
-
-         initimage(output, image->opts, "memory", image->stride_extra);
-         output->input_memory = malloc(size);
-
-         if (output->input_memory != NULL)
-         {
-            output->input_memory_size = size;
-
-            if (png_image_write_to_memory(&image->image, output->input_memory,
-                  &output->input_memory_size, convert_to_8bit, image->buffer+16,
-                  (png_int_32)image->stride, image->colormap))
-            {
-               /* This is also non-fatal but it safes safer to error out anyway:
-                */
-               if (size != output->input_memory_size)
-                  return logerror(image, "memory", ": memory size wrong", "");
-            }
-
-            else
-               return logerror(image, "memory", ": write failed", "");
-         }
-
-         else
-            return logerror(image, "memory", ": out of memory", "");
-      }
-
-      else
-         return logerror(image, "memory", ": get size:", "");
    }
 
    /* 'output' has an initialized temporary image, read this back in and compare
@@ -3439,8 +3356,6 @@ test_one_file(const char *file_name, format_list *formats, png_uint_32 opts,
    int result;
    Image image;
 
-   if (!(opts & NO_RESEED))
-      reseed(); /* ensure that the random numbers don't depend on file order */
    newimage(&image);
    initimage(&image, opts, file_name, stride_extra);
    result = read_one_file(&image);
@@ -3478,7 +3393,7 @@ test_one_file(const char *file_name, format_list *formats, png_uint_32 opts,
 int
 main(int argc, char **argv)
 {
-   png_uint_32 opts = FAST_WRITE | STRICT;
+   png_uint_32 opts = FAST_WRITE;
    format_list formats;
    const char *touch = NULL;
    int log_pass = 0;
@@ -3487,17 +3402,11 @@ main(int argc, char **argv)
    int retval = 0;
    int c;
 
-#if PNG_LIBPNG_VER >= 10700
-      /* This error should not exist in 1.7 or later: */
-      opts |= GBG_ERROR;
-#endif
-
    init_sRGB_to_d();
 #if 0
    init_error_via_linear();
 #endif
    format_init(&formats);
-   reseed(); /* initialize random number seeds */
 
    for (c=1; c<argc; ++c)
    {
@@ -3512,17 +3421,17 @@ main(int argc, char **argv)
       }
       else if (strcmp(arg, "--file") == 0)
 #        ifdef PNG_STDIO_SUPPORTED
-            opts |= USE_FILE;
+            opts |= READ_FILE;
 #        else
-            return SKIP; /* skipped: no support */
+            return 77; /* skipped: no support */
 #        endif
       else if (strcmp(arg, "--memory") == 0)
-         opts &= ~USE_FILE;
+         opts &= ~READ_FILE;
       else if (strcmp(arg, "--stdio") == 0)
 #        ifdef PNG_STDIO_SUPPORTED
             opts |= USE_STDIO;
 #        else
-            return SKIP; /* skipped: no support */
+            return 77; /* skipped: no support */
 #        endif
       else if (strcmp(arg, "--name") == 0)
          opts &= ~USE_STDIO;
@@ -3548,16 +3457,10 @@ main(int argc, char **argv)
          opts &= ~KEEP_GOING;
       else if (strcmp(arg, "--strict") == 0)
          opts |= STRICT;
-      else if (strcmp(arg, "--nostrict") == 0)
-         opts &= ~STRICT;
       else if (strcmp(arg, "--sRGB-16bit") == 0)
          opts |= sRGB_16BIT;
       else if (strcmp(arg, "--linear-16bit") == 0)
          opts &= ~sRGB_16BIT;
-      else if (strcmp(arg, "--noreseed") == 0)
-         opts |= NO_RESEED;
-      else if (strcmp(arg, "--fault-gbg-warning") == 0)
-         opts |= GBG_ERROR;
       else if (strcmp(arg, "--tmpfile") == 0)
       {
          if (c+1 < argc)
@@ -3814,6 +3717,6 @@ int main(void)
 {
    fprintf(stderr, "pngstest: no read support in libpng, test skipped\n");
    /* So the test is skipped: */
-   return SKIP;
+   return 77;
 }
 #endif /* PNG_SIMPLIFIED_READ_SUPPORTED */

contrib/libtests/pngunknown.c

@@ -1,8 +1,8 @@
 
 /* pngunknown.c - test the read side unknown chunk handling
  *
- * Last changed in libpng 1.6.26 [October 20, 2016]
- * Copyright (c) 2015,2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.6.10 [March 6, 2014]
+ * Copyright (c) 2014 Glenn Randers-Pehrson
  * Written by John Cunningham Bowler
  *
  * This code is released under the libpng license.
@@ -30,21 +30,10 @@
 #  include "../../png.h"
 #endif
 
-/* 1.6.1 added support for the configure test harness, which uses 77 to indicate
- * a skipped test, in earlier versions we need to succeed on a skipped test, so:
- */
-#if PNG_LIBPNG_VER >= 10601 && defined(HAVE_CONFIG_H)
-#  define SKIP 77
-#else
-#  define SKIP 0
-#endif
-
-
 /* Since this program tests the ability to change the unknown chunk handling
  * these must be defined:
  */
 #if defined(PNG_SET_UNKNOWN_CHUNKS_SUPPORTED) &&\
-   defined(PNG_STDIO_SUPPORTED) &&\
    defined(PNG_READ_SUPPORTED)
 
 /* One of these must be defined to allow us to find out what happened.  It is
@@ -478,7 +467,7 @@ get_valid(display *d, png_infop info_ptr)
       png_textp text;
       png_uint_32 ntext = png_get_text(d->png_ptr, info_ptr, &text, NULL);
 
-      while (ntext > 0) switch (text[--ntext].compression)
+      while (ntext-- > 0) switch (text[ntext].compression)
       {
          case -1:
             flags |= PNG_INFO_tEXt;
@@ -626,7 +615,7 @@ get_unknown(display *d, png_infop info_ptr, int after_IDAT)
 
    return flags;
 }
-#else /* SAVE_UNKNOWN_CHUNKS */
+#else
 static png_uint_32
 get_unknown(display *d, png_infop info_ptr, int after_IDAT)
    /* Otherwise this will return the cached values set by any user callback */
@@ -645,8 +634,8 @@ get_unknown(display *d, png_infop info_ptr, int after_IDAT)
        * a check to ensure the logic is correct.
        */
 #     error No store support and no user chunk support, this will not work
-#  endif /* READ_USER_CHUNKS */
-#endif /* SAVE_UNKNOWN_CHUNKS */
+#  endif
+#endif
 
 static int
 check(FILE *fp, int argc, const char **argv, png_uint_32p flags/*out*/,
@@ -1012,20 +1001,6 @@ perform_one_test(FILE *fp, int argc, const char **argv,
 
    def = check(fp, argc, argv, flags[1], d, set_callback);
 
-   /* If IDAT is being handled as unknown the image read is skipped and all the
-    * IDATs after the first end up in the end info struct, so in this case add
-    * IDAT to the list of unknowns.  (Do this after 'check' above sets the
-    * chunk_info 'keep' fields.)
-    *
-    * Note that the flag setting has to be in the 'known' field to avoid
-    * triggering the consistency check below and the flag must only be set if
-    * there are multiple IDATs, so if the check above did find an unknown IDAT
-    * after IDAT.
-    */
-   if (chunk_info[0/*IDAT*/].keep != PNG_HANDLE_CHUNK_AS_DEFAULT &&
-       (flags[1][3] & PNG_INFO_IDAT) != 0)
-      flags[0][2] |= PNG_INFO_IDAT;
-
    /* Chunks should either be known or unknown, never both and this should apply
     * whether the chunk is before or after the IDAT (actually, the app can
     * probably change this by swapping the handling after the image, but this
@@ -1270,7 +1245,7 @@ main(void)
    fprintf(stderr,
       " test ignored: no support to find out about unknown chunks\n");
    /* So the test is skipped: */
-   return SKIP;
+   return 77;
 }
 #endif /* READ_USER_CHUNKS || SAVE_UNKNOWN_CHUNKS */
 
@@ -1281,6 +1256,6 @@ main(void)
    fprintf(stderr,
       " test ignored: no support to modify unknown chunk handling\n");
    /* So the test is skipped: */
-   return SKIP;
+   return 77;
 }
 #endif /* SET_UNKNOWN_CHUNKS && READ*/

contrib/libtests/readpng.c

@@ -31,6 +31,13 @@
 #  include "../../png.h"
 #endif
 
+#if PNG_LIBPNG_VER < 10700
+   /* READ_INTERLACING was used instead of READ_DEINTERLACE. */
+#  ifdef PNG_READ_INTERLACING_SUPPORTED
+#     define PNG_READ_DEINTERLACE_SUPPORTED
+#  endif
+#endif
+
 static int
 read_png(FILE *fp)
 {
@@ -62,7 +69,6 @@ read_png(FILE *fp)
    {
       png_size_t rowbytes = png_get_rowbytes(png_ptr, info_ptr);
 
-      /* Failure to initialize these is harmless */
       row = malloc(rowbytes);
       display = malloc(rowbytes);
 
@@ -71,12 +77,12 @@ read_png(FILE *fp)
 
       {
          png_uint_32 height = png_get_image_height(png_ptr, info_ptr);
-#        ifdef PNG_READ_INTERLACING_SUPPORTED
+#        ifdef PNG_READ_DEINTERLACE_SUPPORTED
             int passes = png_set_interlace_handling(png_ptr);
-#        else /* !READ_INTERLACING */
-            int passes = png_get_interlace_type(png_ptr, info_ptr) ==
+#        else
+            int passes = png_get_interlace_type(png_ptr, info_ptr) == 
                PNG_INTERLACE_ADAM7 ? PNG_INTERLACE_ADAM7_PASSES : 1;
-#        endif /* !READ_INTERLACING */
+#        endif
          int pass;
 
          png_start_read_image(png_ptr);
@@ -85,10 +91,10 @@ read_png(FILE *fp)
          {
             png_uint_32 y = height;
 
-#           ifndef PNG_READ_INTERLACING_SUPPORTED
+#           ifndef PNG_READ_DEINTERLACE_SUPPORTED
                if (passes == PNG_INTERLACE_ADAM7_PASSES)
                   y = PNG_PASS_ROWS(y, pass);
-#           endif /* READ_INTERLACING */
+#           endif
 
             /* NOTE: this trashes the row each time; interlace handling won't
              * work, but this avoids memory thrashing for speed testing.

contrib/libtests/timepng.c

@@ -1,8 +1,8 @@
 /* timepng.c
  *
- * Copyright (c) 2013,2016 John Cunningham Bowler
+ * Copyright (c) 2013 John Cunningham Bowler
  *
- * Last changed in libpng 1.6.22 [May 26, 2016]
+ * Last changed in libpng 1.6.1 [March 28, 2013]
  *
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
@@ -10,17 +10,15 @@
  *
  * Load an arbitrary number of PNG files (from the command line, or, if there
  * are no arguments on the command line, from stdin) then run a time test by
- * reading each file by row or by image (possibly with transforms in the latter
- * case).  The only output is a time as a floating point number of seconds with
- * 9 decimal digits.
+ * reading each file by row.  The test does nothing with the read result and
+ * does no transforms.  The only output is a time as a floating point number of
+ * seconds with 9 decimal digits.
  */
 #define _POSIX_C_SOURCE 199309L /* for clock_gettime */
 
 #include <stdlib.h>
 #include <stdio.h>
 #include <string.h>
-#include <errno.h>
-#include <limits.h>
 
 #include <time.h>
 
@@ -37,73 +35,36 @@
 #  include "../../png.h"
 #endif
 
-/* The following is to support direct compilation of this file as C++ */
-#ifdef __cplusplus
-#  define voidcast(type, value) static_cast<type>(value)
-#else
-#  define voidcast(type, value) (value)
-#endif /* __cplusplus */
-
-/* 'CLOCK_PROCESS_CPUTIME_ID' is one of the clock timers for clock_gettime.  It
- * need not be supported even when clock_gettime is available.  It returns the
- * 'CPU' time the process has consumed.  'CPU' time is assumed to include time
- * when the CPU is actually blocked by a pending cache fill but not time
- * waiting for page faults.  The attempt is to get a measure of the actual time
- * the implementation takes to read a PNG ignoring the potentially very large IO
- * overhead.
- */
-#if defined (CLOCK_PROCESS_CPUTIME_ID) && defined(PNG_STDIO_SUPPORTED) &&\
-    defined(PNG_EASY_ACCESS_SUPPORTED) &&\
-    (PNG_LIBPNG_VER >= 10700 ? defined(PNG_READ_PNG_SUPPORTED) :\
-     defined (PNG_SEQUENTIAL_READ_SUPPORTED) &&\
-     defined(PNG_INFO_IMAGE_SUPPORTED))
-
-typedef struct
+static int read_png(FILE *fp)
 {
-   FILE *input;
-   FILE *output;
-}  io_data;
-
-static PNG_CALLBACK(void, read_and_copy,
-      (png_structp png_ptr, png_bytep buffer, png_size_t cb))
-{
-   io_data *io = (io_data*)png_get_io_ptr(png_ptr);
-
-   if (fread(buffer, cb, 1, io->input) != 1)
-      png_error(png_ptr, strerror(errno));
-
-   if (fwrite(buffer, cb, 1, io->output) != 1)
-   {
-      perror("temporary file");
-      fprintf(stderr, "temporary file PNG write failed\n");
-      exit(1);
-   }
-}
-
-static void read_by_row(png_structp png_ptr, png_infop info_ptr,
-      FILE *write_ptr, FILE *read_ptr)
-{
-   /* These don't get freed on error, this is fine; the program immediately
-    * exits.
-    */
+   png_structp png_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING,0,0,0);
+   png_infop info_ptr = NULL;
    png_bytep row = NULL, display = NULL;
-   io_data io_copy;
 
-   if (write_ptr != NULL)
+   if (png_ptr == NULL)
+      return 0;
+
+   if (setjmp(png_jmpbuf(png_ptr)))
    {
-      /* Set up for a copy to the temporary file: */
-      io_copy.input = read_ptr;
-      io_copy.output = write_ptr;
-      png_set_read_fn(png_ptr, &io_copy, read_and_copy);
+      png_destroy_read_struct(&png_ptr, &info_ptr, NULL);
+      if (row != NULL) free(row);
+      if (display != NULL) free(display);
+      return 0;
    }
 
+   png_init_io(png_ptr, fp);
+
+   info_ptr = png_create_info_struct(png_ptr);
+   if (info_ptr == NULL)
+      png_error(png_ptr, "OOM allocating info structure");
+
    png_read_info(png_ptr, info_ptr);
 
    {
       png_size_t rowbytes = png_get_rowbytes(png_ptr, info_ptr);
 
-      row = voidcast(png_bytep,malloc(rowbytes));
-      display = voidcast(png_bytep,malloc(rowbytes));
+      row = malloc(rowbytes);
+      display = malloc(rowbytes);
 
       if (row == NULL || display == NULL)
          png_error(png_ptr, "OOM allocating row buffers");
@@ -120,8 +81,7 @@ static void read_by_row(png_structp png_ptr, png_infop info_ptr,
             png_uint_32 y = height;
 
             /* NOTE: this trashes the row each time; interlace handling won't
-             * work, but this avoids memory thrashing for speed testing and is
-             * somewhat representative of an application that works row-by-row.
+             * work, but this avoids memory thrashing for speed testing.
              */
             while (y-- > 0)
                png_read_row(png_ptr, row, display);
@@ -131,51 +91,9 @@ static void read_by_row(png_structp png_ptr, png_infop info_ptr,
 
    /* Make sure to read to the end of the file: */
    png_read_end(png_ptr, info_ptr);
-
-   /* Free this up: */
+   png_destroy_read_struct(&png_ptr, &info_ptr, NULL);
    free(row);
    free(display);
-}
-
-static PNG_CALLBACK(void, no_warnings, (png_structp png_ptr,
-         png_const_charp warning))
-{
-   (void)png_ptr;
-   (void)warning;
-}
-
-static int read_png(FILE *fp, png_int_32 transforms, FILE *write_file)
-{
-   png_structp png_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING,0,0,
-         no_warnings);
-   png_infop info_ptr = NULL;
-
-   if (png_ptr == NULL)
-      return 0;
-
-   if (setjmp(png_jmpbuf(png_ptr)))
-   {
-      png_destroy_read_struct(&png_ptr, &info_ptr, NULL);
-      return 0;
-   }
-
-#  ifdef PNG_BENIGN_ERRORS_SUPPORTED
-      png_set_benign_errors(png_ptr, 1/*allowed*/);
-#  endif
-   png_init_io(png_ptr, fp);
-
-   info_ptr = png_create_info_struct(png_ptr);
-
-   if (info_ptr == NULL)
-      png_error(png_ptr, "OOM allocating info structure");
-
-   if (transforms < 0)
-      read_by_row(png_ptr, info_ptr, write_file, fp);
-
-   else
-      png_read_png(png_ptr, info_ptr, transforms, NULL/*params*/);
-
-   png_destroy_read_struct(&png_ptr, &info_ptr, NULL);
    return 1;
 }
 
@@ -190,7 +108,7 @@ static int mytime(struct timespec *t)
    return 0;
 }
 
-static int perform_one_test(FILE *fp, int nfiles, png_int_32 transforms)
+static int perform_one_test(FILE *fp, int nfiles)
 {
    int i;
    struct timespec before, after;
@@ -202,7 +120,7 @@ static int perform_one_test(FILE *fp, int nfiles, png_int_32 transforms)
    {
       for (i=0; i<nfiles; ++i)
       {
-         if (read_png(fp, transforms, NULL/*write*/))
+         if (read_png(fp))
          {
             if (ferror(fp))
             {
@@ -266,343 +184,120 @@ static int add_one_file(FILE *fp, char *name)
 
    if (ip != NULL)
    {
-      /* Read the file using libpng; this detects errors and also deals with
-       * files which contain data beyond the end of the file.
-       */
-      int ok = 0;
-      fpos_t pos;
-
-      if (fgetpos(fp, &pos))
+      int ch;
+      for (;;)
       {
-         /* Fatal error reading the start: */
-         perror("temporary file");
-         fprintf(stderr, "temporary file fgetpos error\n");
-         exit(1);
+         ch = getc(ip);
+         if (ch == EOF) break;
+         putc(ch, fp);
       }
 
-      if (read_png(ip, -1/*by row*/, fp/*output*/))
+      if (ferror(ip))
       {
-         if (ferror(ip))
-         {
-            perror(name);
-            fprintf(stderr, "%s: read error\n", name);
-         }
-
-         else
-            ok = 1; /* read ok */
+         perror(name);
+         fprintf(stderr, "%s: read error\n", name);
+         return 0;
       }
 
-      else
-         fprintf(stderr, "%s: file not added\n", name);
-
       (void)fclose(ip);
 
-      /* An error in the output is fatal; exit immediately: */
       if (ferror(fp))
       {
          perror("temporary file");
          fprintf(stderr, "temporary file write error\n");
-         exit(1);
-      }
-
-      if (ok)
-         return 1;
-
-      /* Did not read the file successfully, simply rewind the temporary
-       * file.  This must happen after the ferror check above to avoid clearing
-       * the error.
-       */
-      if (fsetpos(fp, &pos))
-      {
-         perror("temporary file");
-         fprintf(stderr, "temporary file fsetpos error\n");
-         exit(1);
+         return 0;
       }
    }
 
    else
    {
-      /* file open error: */
       perror(name);
       fprintf(stderr, "%s: open failed\n", name);
+      return 0;
    }
 
-   return 0; /* file not added */
-}
-
-static void
-usage(FILE *fp)
-{
-   if (fp != NULL) fclose(fp);
-
-   fprintf(stderr,
-"Usage:\n"
-" timepng --assemble <assembly> {files}\n"
-"  Read the files into <assembly>, output the count.  Options are ignored.\n"
-" timepng --dissemble <assembly> <count> [options]\n"
-"  Time <count> files from <assembly>, additional files may not be given.\n"
-" Otherwise:\n"
-"  Read the files into a temporary file and time the decode\n"
-"Transforms:\n"
-"  --by-image: read by image with png_read_png\n"
-"  --<transform>: implies by-image, use PNG_TRANSFORM_<transform>\n"
-"  Otherwise: read by row using png_read_row (to a single row buffer)\n"
-   /* ISO C90 string length max 509 */);fprintf(stderr,
-"{files}:\n"
-"  PNG files to copy into the assembly and time.  Invalid files are skipped\n"
-"  with appropriate error messages.  If no files are given the list of files\n"
-"  is read from stdin with each file name terminated by a newline\n"
-"Output:\n"
-"  For --assemble the output is the name of the assembly file followed by the\n"
-"  count of the files it contains; the arguments for --dissemble.  Otherwise\n"
-"  the output is the total decode time in seconds.\n");
-
-   exit(99);
+   return 1;
 }
 
 int main(int argc, char **argv)
 {
    int ok = 0;
-   int err = 0;
-   int nfiles = 0;
-   int transforms = -1; /* by row */
-   const char *assembly = NULL;
-   FILE *fp;
+   FILE *fp = tmpfile();
 
-   if (argc > 2 && strcmp(argv[1], "--assemble") == 0)
+   if (fp != NULL)
    {
-      /* Just build the test file, argv[2] is the file name. */
-      assembly = argv[2];
-      fp = fopen(assembly, "wb");
-      if (fp == NULL)
+      int err = 0;
+      int nfiles = 0;
+
+      if (argc > 1)
       {
-         perror(assembly);
-         fprintf(stderr, "timepng --assemble %s: could not open for write\n",
-               assembly);
-         usage(NULL);
-      }
+         int i;
 
-      argv += 2;
-      argc -= 2;
-   }
-
-   else if (argc > 3 && strcmp(argv[1], "--dissemble") == 0)
-   {
-      fp = fopen(argv[2], "rb");
-
-      if (fp == NULL)
-      {
-         perror(argv[2]);
-         fprintf(stderr, "timepng --dissemble %s: could not open for read\n",
-               argv[2]);
-         usage(NULL);
-      }
-
-      nfiles = atoi(argv[3]);
-      if (nfiles <= 0)
-      {
-         fprintf(stderr,
-               "timepng --dissemble <file> <count>: %s is not a count\n",
-               argv[3]);
-         exit(99);
-      }
-#ifdef __COVERITY__
-      else
-      {
-         nfiles &= PNG_UINT_31_MAX;
-      }
-#endif
-
-      argv += 3;
-      argc -= 3;
-   }
-
-   else /* Else use a temporary file */
-   {
-#ifndef __COVERITY__
-      fp = tmpfile();
-#else
-      /* Experimental. Coverity says tmpfile() is insecure because it
-       * generates predictable names.
-       *
-       * It is possible to satisfy Coverity by using mkstemp(); however,
-       * any platform supporting mkstemp() undoubtedly has a secure tmpfile()
-       * implementation as well, and doesn't need the fix.  Note that
-       * the fix won't work on platforms that don't support mkstemp().
-       *
-       * https://www.securecoding.cert.org/confluence/display/c/
-       * FIO21-C.+Do+not+create+temporary+files+in+shared+directories
-       * says that most historic implementations of tmpfile() provide
-       * only a limited number of possible temporary file names
-       * (usually 26) before file names are recycled. That article also
-       * provides a secure solution that unfortunately depends upon mkstemp().
-       */
-      char tmpfile[] = "timepng-XXXXXX";
-      int filedes;
-      umask(0177);
-      filedes = mkstemp(tmpfile);
-      if (filedes < 0)
-        fp = NULL;
-      else
-      {
-        fp = fdopen(filedes,"w+");
-        /* Hide the filename immediately and ensure that the file does
-         * not exist after the program ends
-         */
-        (void) unlink(tmpfile);
-      }
-#endif
-
-      if (fp == NULL)
-      {
-         perror("tmpfile");
-         fprintf(stderr, "timepng: could not open the temporary file\n");
-         exit(1); /* not a user error */
-      }
-   }
-
-   /* Handle the transforms: */
-   while (argc > 1 && argv[1][0] == '-' && argv[1][1] == '-')
-   {
-      const char *opt = *++argv + 2;
-
-      --argc;
-
-      /* Transforms turn on the by-image processing and maybe set some
-       * transforms:
-       */
-      if (transforms == -1)
-         transforms = PNG_TRANSFORM_IDENTITY;
-
-      if (strcmp(opt, "by-image") == 0)
-      {
-         /* handled above */
-      }
-
-#        define OPT(name) else if (strcmp(opt, #name) == 0)\
-         transforms |= PNG_TRANSFORM_ ## name
-
-      OPT(STRIP_16);
-      OPT(STRIP_ALPHA);
-      OPT(PACKING);
-      OPT(PACKSWAP);
-      OPT(EXPAND);
-      OPT(INVERT_MONO);
-      OPT(SHIFT);
-      OPT(BGR);
-      OPT(SWAP_ALPHA);
-      OPT(SWAP_ENDIAN);
-      OPT(INVERT_ALPHA);
-      OPT(STRIP_FILLER);
-      OPT(STRIP_FILLER_BEFORE);
-      OPT(STRIP_FILLER_AFTER);
-      OPT(GRAY_TO_RGB);
-      OPT(EXPAND_16);
-      OPT(SCALE_16);
-
-      else
-      {
-         fprintf(stderr, "timepng %s: unrecognized transform\n", opt);
-         usage(fp);
-      }
-   }
-
-   /* Handle the files: */
-   if (argc > 1 && nfiles > 0)
-      usage(fp); /* Additional files not valid with --dissemble */
-
-   else if (argc > 1)
-   {
-      int i;
-
-      for (i=1; i<argc; ++i)
-      {
-         if (nfiles == INT_MAX)
+         for (i=1; i<argc; ++i)
          {
-            fprintf(stderr, "%s: skipped, too many files\n", argv[i]);
-            break;
-         }
+            if (add_one_file(fp, argv[i]))
+               ++nfiles;
 
-         else if (add_one_file(fp, argv[i]))
-            ++nfiles;
-      }
-   }
-
-   else if (nfiles == 0) /* Read from stdin withoout --dissemble */
-   {
-      char filename[FILENAME_MAX+1];
-
-      while (fgets(filename, FILENAME_MAX+1, stdin))
-      {
-         size_t len = strlen(filename);
-
-         if (filename[len-1] == '\n')
-         {
-            filename[len-1] = 0;
-            if (nfiles == INT_MAX)
+            else
             {
-               fprintf(stderr, "%s: skipped, too many files\n", filename);
+               err = 1;
                break;
             }
-
-            else if (add_one_file(fp, filename))
-               ++nfiles;
-         }
-
-         else
-         {
-            fprintf(stderr, "timepng: file name too long: ...%s\n",
-               filename+len-32);
-            err = 1;
-            break;
          }
       }
 
-      if (ferror(stdin))
+      else
       {
-         fprintf(stderr, "timepng: stdin: read error\n");
-         err = 1;
-      }
-   }
+         char filename[FILENAME_MAX+1];
 
-   /* Perform the test, or produce the --assemble output: */
-   if (!err)
-   {
-      if (nfiles > 0)
-      {
-         if (assembly != NULL)
+         while (fgets(filename, FILENAME_MAX+1, stdin))
          {
-            if (fflush(fp) && !ferror(fp) && fclose(fp))
+            size_t len = strlen(filename);
+
+            if (filename[len-1] == '\n')
             {
-               perror(assembly);
-               fprintf(stderr, "%s: close failed\n", assembly);
+               filename[len-1] = 0;
+               if (add_one_file(fp, filename))
+                  ++nfiles;
+
+               else
+               {
+                  err = 1;
+                  break;
+               }
             }
 
             else
             {
-               printf("%s %d\n", assembly, nfiles);
-               fflush(stdout);
-               ok = !ferror(stdout);
+               fprintf(stderr, "timepng: truncated file name ...%s\n",
+                  filename+len-32);
+               err = 1;
+               break;
             }
          }
 
-         else
+         if (ferror(stdin))
          {
-            ok = perform_one_test(fp, nfiles, transforms);
-            (void)fclose(fp);
+            fprintf(stderr, "timepng: stdin: read error\n");
+            err = 1;
          }
       }
 
-      else
-         usage(fp);
+      if (!err)
+      {
+         if (nfiles > 0)
+            ok = perform_one_test(fp, nfiles);
+
+         else
+            fprintf(stderr, "usage: timepng {files} or ls files | timepng\n");
+      }
+
+      (void)fclose(fp);
    }
 
    else
-      (void)fclose(fp);
+      fprintf(stderr, "timepng: could not open temporary file\n");
 
    /* Exit code 0 on success. */
    return ok == 0;
 }
-#else /* !sufficient support */
-int main(void) { return 77; }
-#endif /* !sufficient support */

contrib/mips-msa/README

@@ -1,83 +0,0 @@
-OPERATING SYSTEM SPECIFIC MIPS MSA DETECTION
---------------------------------------------
-
-Detection of the ability to execute MIPS MSA on an MIPS processor requires
-operating system support.  (The information is not available in user mode.)
-
-HOW TO USE THIS
----------------
-
-This directory contains C code fragments that can be included in mips/mips_init.c
-by setting the macro PNG_MIPS_MSA_FILE to the file name in "" or <> at build
-time.  This setting is not recorded in pnglibconf.h and can be changed simply by
-rebuilding mips/msa_init.o with the required macro definition.
-
-For any of this code to be used the MIPS MSA code must be enabled and run time
-checks must be supported.  I.e.:
-
-#if PNG_MIPS_MSA_OPT > 0
-#ifdef PNG_MIPS_MSA_CHECK_SUPPORTED
-
-This is done in a 'configure' build by passing configure the argument:
-
-   --enable-mips-msa=check
-
-Apart from the basic Linux implementation in contrib/mips-msa/linux.c this code
-is unsupported.  That means that it is not even compiled on a regular basis and
-may be broken in any given minor release.
-
-FILE FORMAT
------------
-
-Each file documents its testing status as of the last time it was tested (which
-may have been a long time ago):
-
-STATUS: one of:
-   SUPPORTED: This indicates that the file is included in the regularly
-         performed test builds and bugs are fixed when discovered.
-   COMPILED: This indicates that the code did compile at least once.  See the
-         more detailed description for the extent to which the result was
-         successful.
-   TESTED: This means the code was fully compiled into the libpng test programs
-         and these were run at least once.
-
-BUG REPORTS: an email address to which to send reports of problems
-
-The file is a fragment of C code. It should not define any 'extern' symbols;
-everything should be static.  It must define the function:
-
-static int png_have_msa(png_structp png_ptr);
-
-That function must return 1 if MIPS MSA instructions are supported, 0 if not.
-It must not execute png_error unless it detects a bug.  A png_error will prevent
-the reading of the PNG and in the future, writing too.
-
-BUG REPORTS
------------
-
-If you mail a bug report for any file that is not SUPPORTED there may only be
-limited response.  Consider fixing it and sending a patch to fix the problem -
-this is more likely to result in action.
-
-CONTRIBUTIONS
--------------
-
-You may send contributions of new implementations to
-png-mng-implement@sourceforge.net.  Please write code in strict C90 C where
-possible.  Obviously OS dependencies are to be expected.  If you submit code you
-must have the authors permission and it must have a license that is acceptable
-to the current maintainer; in particular that license must permit modification
-and redistribution.
-
-Please try to make the contribution a single file and give the file a clear and
-unambiguous name that identifies the target OS.  If multiple files really are
-required put them all in a sub-directory.
-
-You must also be prepared to handle bug reports from users of the code, either
-by joining the png-mng-implement mailing list or by providing an email for the
-"BUG REPORTS" entry or both.  Please make sure that the header of the file
-contains the STATUS and BUG REPORTS fields as above.
-
-Please list the OS requirements as precisely as possible.  Ideally you should
-also list the environment in which the code has been tested and certainly list
-any environments where you suspect it might not work.

contrib/mips-msa/linux.c

@@ -1,64 +0,0 @@
-/* contrib/mips-msa/linux.c
- *
- * Copyright (c) 2016 Glenn Randers-Pehrson
- * Written by Mandar Sahastrabuddhe, 2016.
- * Last changed in libpng 1.6.25beta03 [August 29, 2016]
- *
- * This code is released under the libpng license.
- * For conditions of distribution and use, see the disclaimer
- * and license in png.h
- *
- * SEE contrib/mips-msa/README before reporting bugs
- *
- * STATUS: SUPPORTED
- * BUG REPORTS: png-mng-implement@sourceforge.net
- *
- * png_have_msa implemented for Linux by reading the widely available
- * pseudo-file /proc/cpuinfo.
- *
- * This code is strict ANSI-C and is probably moderately portable; it does
- * however use <stdio.h> and it assumes that /proc/cpuinfo is never localized.
- */
-
-#include <stdio.h>
-#include <string.h>
-#include <stdlib.h>
-
-static int
-png_have_msa(png_structp png_ptr)
-{
-   FILE *f = fopen("/proc/cpuinfo", "rb");
-
-   char *string = "msa";
-   char word[10];
-
-   if (f != NULL)
-   {
-      while(!feof(f))
-      {
-         int ch = fgetc(f);
-         static int i = 0;
-
-         while(!(ch <= 32))
-         {
-            word[i++] = ch;
-            ch = fgetc(f);
-         }
-
-         int val = strcmp(string, word);
-
-         if (val == 0)
-            return 1;
-
-         i = 0;
-         memset(word, 0, 10);
-      }
-
-      fclose(f);
-   }
-#ifdef PNG_WARNINGS_SUPPORTED
-   else
-      png_warning(png_ptr, "/proc/cpuinfo open failed");
-#endif
-   return 0;
-}

contrib/neon/arm_init.c

@@ -1,9 +1,9 @@
 
 /* arm_init.c - NEON optimised filter functions
  *
- * Copyright (c) 2014,2016 Glenn Randers-Pehrson
+ * Copyright (c) 2014 Glenn Randers-Pehrson
  * Written by Mans Rullgard, 2011.
- * Last changed in libpng 1.6.22 [May 26, 2016]
+ * Last changed in libpng 1.6.16 [December 22, 2014]
  *
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
@@ -14,7 +14,8 @@
  */
 #define _POSIX_SOURCE 1
 
-#include "../pngpriv.h"
+#include "../../pngpriv.h"
+#define PNG_SRC_FILE PNG_SRC_FILE_arm_arm_init
 
 #ifdef PNG_READ_SUPPORTED
 
@@ -66,7 +67,6 @@ png_init_filter_functions_neon(png_structp pp, unsigned int bpp)
     * wrong order of the 'ON' and 'default' cases.  UNSET now defaults to OFF,
     * as documented in png.h
     */
-   png_debug(1, "in png_init_filter_functions_neon");
 #ifdef PNG_ARM_NEON_API_SUPPORTED
    switch ((pp->options >> PNG_ARM_NEON) & 3)
    {

contrib/neon/extensions.ac

@@ -0,0 +1,10 @@
+# To just this extension either copy the 'sinclude' line and the setting of
+# libpng_automake_extensions into contrib/extensions.am or link/copy the whole
+# file there.  Do the same with the associated file extensions.am
+#
+# Include the ARM NEON extension:
+sinclude([contrib/neon/neon.ac])
+#
+# If any of the extensions require automake support (this is likely) automake
+# must include the file contrib/extensions.am
+libpng_automake_extensions="yes"

contrib/neon/extensions.am

@@ -0,0 +1,5 @@
+# To just this extension either copy the 'include' line into
+# contrib/extensions.am or link/copy the whole file there.  Do the same with the
+# associated file extensions.ac
+# Include the ARM NEON extension:
+include %reldir%/neon/neon.am

contrib/neon/filter_neon.S

@@ -3,7 +3,7 @@
  *
  * Copyright (c) 2014 Glenn Randers-Pehrson
  * Written by Mans Rullgard, 2011.
- * Last changed in libpng 1.6.16 [December 22, 2014]
+ * Last changed in libpng 1.6.16 [(PENDING RELEASE)]
  *
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
@@ -14,7 +14,7 @@
  * definitions (or not) of PNG_ARM_NEON_OPT and PNG_ARM_NEON_IMPLEMENTATION.
  */
 #define PNG_VERSION_INFO_ONLY
-#include "../pngpriv.h"
+#include "../../pngpriv.h"
 
 #if defined(__linux__) && defined(__ELF__)
 .section .note.GNU-stack,"",%progbits /* mark stack as non-executable */

contrib/neon/filter_neon_intrinsics.c

@@ -1,18 +1,19 @@
 
 /* filter_neon_intrinsics.c - NEON optimised filter functions
  *
- * Copyright (c) 2014,2016 Glenn Randers-Pehrson
+ * Copyright (c) 2014 Glenn Randers-Pehrson
  * Written by James Yu <james.yu at linaro.org>, October 2013.
  * Based on filter_neon.S, written by Mans Rullgard, 2011.
  *
- * Last changed in libpng 1.6.22 [May 26, 2016]
+ * Last changed in libpng 1.6.16 [December 22, 2014]
  *
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
  * and license in png.h
  */
 
-#include "../pngpriv.h"
+#include "../../pngpriv.h"
+#define PNG_SRC_FILE PNG_SRC_FILE_arm_filter_neon_intrinsics
 
 #ifdef PNG_READ_SUPPORTED
 
@@ -40,15 +41,13 @@
 #if PNG_ARM_NEON_OPT > 0
 
 void
-png_read_filter_row_up_neon(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev_row)
+png_read_filter_row_up_neon(png_alloc_size_t rowbytes, unsigned int bpp,
+   png_bytep row, png_const_bytep prev_row)
 {
    png_bytep rp = row;
-   png_bytep rp_stop = row + row_info->rowbytes;
+   png_bytep rp_stop = row + rowbytes;
    png_const_bytep pp = prev_row;
 
-   png_debug(1, "in png_read_filter_row_up_neon");
-
    for (; rp < rp_stop; rp += 16, pp += 16)
    {
       uint8x16_t qrp, qpp;
@@ -58,14 +57,15 @@ png_read_filter_row_up_neon(png_row_infop row_info, png_bytep row,
       qrp = vaddq_u8(qrp, qpp);
       vst1q_u8(rp, qrp);
    }
+
+   PNG_UNUSED(bpp)
 }
 
 void
-png_read_filter_row_sub3_neon(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev_row)
+png_read_filter_row_sub3_neon(png_alloc_size_t rowbytes, unsigned int bpp, png_bytep row, png_const_bytep prev_row)
 {
    png_bytep rp = row;
-   png_bytep rp_stop = row + row_info->rowbytes;
+   png_bytep rp_stop = row + rowbytes;
 
    uint8x16_t vtmp = vld1q_u8(rp);
    uint8x8x2_t *vrpt = png_ptr(uint8x8x2_t, &vtmp);
@@ -74,8 +74,6 @@ png_read_filter_row_sub3_neon(png_row_infop row_info, png_bytep row,
    uint8x8x4_t vdest;
    vdest.val[3] = vdup_n_u8(0);
 
-   png_debug(1, "in png_read_filter_row_sub3_neon");
-
    for (; rp < rp_stop;)
    {
       uint8x8_t vtmp1, vtmp2;
@@ -104,21 +102,20 @@ png_read_filter_row_sub3_neon(png_row_infop row_info, png_bytep row,
       rp += 3;
    }
 
+   PNG_UNUSED(bpp)
    PNG_UNUSED(prev_row)
 }
 
 void
-png_read_filter_row_sub4_neon(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev_row)
+png_read_filter_row_sub4_neon(png_alloc_size_t rowbytes, unsigned int bpp,
+   png_bytep row, png_const_bytep prev_row)
 {
    png_bytep rp = row;
-   png_bytep rp_stop = row + row_info->rowbytes;
+   png_bytep rp_stop = row + rowbytes;
 
    uint8x8x4_t vdest;
    vdest.val[3] = vdup_n_u8(0);
 
-   png_debug(1, "in png_read_filter_row_sub4_neon");
-
    for (; rp < rp_stop; rp += 16)
    {
       uint32x2x4_t vtmp = vld4_u32(png_ptr(uint32_t,rp));
@@ -133,16 +130,17 @@ png_read_filter_row_sub4_neon(png_row_infop row_info, png_bytep row,
       vst4_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2x4_t,&vdest), 0);
    }
 
+   PNG_UNUSED(bpp)
    PNG_UNUSED(prev_row)
 }
 
 void
-png_read_filter_row_avg3_neon(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev_row)
+png_read_filter_row_avg3_neon(png_alloc_size_t rowbytes, unsigned int bpp,
+   png_bytep row, png_const_bytep prev_row)
 {
    png_bytep rp = row;
    png_const_bytep pp = prev_row;
-   png_bytep rp_stop = row + row_info->rowbytes;
+   png_bytep rp_stop = row + rowbytes;
 
    uint8x16_t vtmp;
    uint8x8x2_t *vrpt;
@@ -154,8 +152,6 @@ png_read_filter_row_avg3_neon(png_row_infop row_info, png_bytep row,
    vrpt = png_ptr(uint8x8x2_t,&vtmp);
    vrp = *vrpt;
 
-   png_debug(1, "in png_read_filter_row_avg3_neon");
-
    for (; rp < rp_stop; pp += 12)
    {
       uint8x8_t vtmp1, vtmp2, vtmp3;
@@ -202,21 +198,21 @@ png_read_filter_row_avg3_neon(png_row_infop row_info, png_bytep row,
       vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[3]), 0);
       rp += 3;
    }
+
+   PNG_UNUSED(bpp)
 }
 
 void
-png_read_filter_row_avg4_neon(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev_row)
+png_read_filter_row_avg4_neon(png_alloc_size_t rowbytes, unsigned int bpp,
+   png_bytep row, png_const_bytep prev_row)
 {
    png_bytep rp = row;
-   png_bytep rp_stop = row + row_info->rowbytes;
+   png_bytep rp_stop = row + rowbytes;
    png_const_bytep pp = prev_row;
 
    uint8x8x4_t vdest;
    vdest.val[3] = vdup_n_u8(0);
 
-   png_debug(1, "in png_read_filter_row_avg4_neon");
-
    for (; rp < rp_stop; rp += 16, pp += 16)
    {
       uint32x2x4_t vtmp;
@@ -242,6 +238,8 @@ png_read_filter_row_avg4_neon(png_row_infop row_info, png_bytep row,
 
       vst4_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2x4_t,&vdest), 0);
    }
+
+   PNG_UNUSED(bpp)
 }
 
 static uint8x8_t
@@ -272,12 +270,12 @@ paeth(uint8x8_t a, uint8x8_t b, uint8x8_t c)
 }
 
 void
-png_read_filter_row_paeth3_neon(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev_row)
+png_read_filter_row_paeth3_neon(png_alloc_size_t rowbytes, unsigned int bpp,
+   png_bytep row, png_const_bytep prev_row)
 {
    png_bytep rp = row;
    png_const_bytep pp = prev_row;
-   png_bytep rp_stop = row + row_info->rowbytes;
+   png_bytep rp_stop = row + rowbytes;
 
    uint8x16_t vtmp;
    uint8x8x2_t *vrpt;
@@ -290,8 +288,6 @@ png_read_filter_row_paeth3_neon(png_row_infop row_info, png_bytep row,
    vrpt = png_ptr(uint8x8x2_t,&vtmp);
    vrp = *vrpt;
 
-   png_debug(1, "in png_read_filter_row_paeth3_neon");
-
    for (; rp < rp_stop; pp += 12)
    {
       uint8x8x2_t *vppt;
@@ -337,22 +333,22 @@ png_read_filter_row_paeth3_neon(png_row_infop row_info, png_bytep row,
       vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[3]), 0);
       rp += 3;
    }
+
+   PNG_UNUSED(bpp)
 }
 
 void
-png_read_filter_row_paeth4_neon(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev_row)
+png_read_filter_row_paeth4_neon(png_alloc_size_t rowbytes, unsigned int bpp,
+   png_bytep row, png_const_bytep prev_row)
 {
    png_bytep rp = row;
-   png_bytep rp_stop = row + row_info->rowbytes;
+   png_bytep rp_stop = row + rowbytes;
    png_const_bytep pp = prev_row;
 
    uint8x8_t vlast = vdup_n_u8(0);
    uint8x8x4_t vdest;
    vdest.val[3] = vdup_n_u8(0);
 
-   png_debug(1, "in png_read_filter_row_paeth4_neon");
-
    for (; rp < rp_stop; rp += 16, pp += 16)
    {
       uint32x2x4_t vtmp;
@@ -380,8 +376,9 @@ png_read_filter_row_paeth4_neon(png_row_infop row_info, png_bytep row,
 
       vst4_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2x4_t,&vdest), 0);
    }
-}
 
+   PNG_UNUSED(bpp)
+}
 #endif /* PNG_ARM_NEON_OPT > 0 */
 #endif /* PNG_ARM_NEON_IMPLEMENTATION == 1 (intrinsics) */
 #endif /* READ */

contrib/neon/neon.ac

@@ -0,0 +1,47 @@
+# LIBPNG EXTENSION: NEON
+#
+# ARM NEON (SIMD) support.
+#
+# This file contains configure.ac additions to libpng configure to enable
+# building of the ARM NEON filter optimizations.
+#
+AC_ARG_ENABLE([arm-neon],
+   AS_HELP_STRING([[[--enable-arm-neon]]],
+      [Enable ARM NEON optimizations: =no/off, check, api, yes/on:]
+      [no/off: disable the optimizations; check: use internal checking code]
+      [(deprecated and poorly supported); api: disable by default, enable by]
+      [a call to png_set_option; yes/on: turn on unconditionally.]
+      [If not specified: determined by the compiler.]),
+   [case "$enableval" in
+      no|off)
+         # disable the default enabling on __ARM_NEON__ systems:
+         AC_DEFINE([PNG_ARM_NEON_OPT], [0],
+                   [Disable ARM Neon optimizations])
+         # Prevent inclusion of the assembler files below:
+         enable_arm_neon=no;;
+      check)
+         AC_DEFINE([PNG_ARM_NEON_CHECK_SUPPORTED], [],
+                   [Check for ARM Neon support at run-time]);;
+      api)
+         AC_DEFINE([PNG_ARM_NEON_API_SUPPORTED], [],
+                   [Turn on ARM Neon optimizations at run-time]);;
+      yes|on)
+         AC_DEFINE([PNG_ARM_NEON_OPT], [2],
+                   [Enable ARM Neon optimizations])
+         AC_MSG_WARN([--enable-arm-neon: please specify 'check' or 'api', if]
+            [you want the optimizations unconditionally pass -mfpu=neon]
+            [to the compiler.]);;
+      *)
+         AC_MSG_ERROR([--enable-arm-neon=${enable_arm_neon}: invalid value])
+   esac])
+
+# Add ARM specific files to all builds where the host_cpu is arm ('arm*') or
+# where ARM optimizations were explicitly requested (this allows a fallback if a
+# future host CPU does not match 'arm*')
+
+AM_CONDITIONAL([PNG_ARM_NEON],
+   [test "$enable_arm_neon" != 'no' &&
+    case "$host_cpu" in
+      arm*|aarch64*) :;;
+      *)    test "$enable_arm_neon" != '';;
+    esac])

contrib/neon/neon.am

@@ -0,0 +1,27 @@
+# contrib/neon/neon.am
+#
+# This is an automake fragment included by the top-level Makefile.am if
+# contrib/neon/neon.ac is included in configure 
+#
+# It builds the source files in this directory into ARM hardware support for
+# libpng.
+#
+if PNG_ARM_NEON
+libpng@PNGLIB_MAJOR@@PNGLIB_MINOR@_la_SOURCES += %D%/arm_init.c\
+	%D%/filter_neon.S %D%/filter_neon_intrinsics.c
+
+DFA_EXTENSION += ${srcdir}/%D%/neon.dfa
+
+# This causes pngpriv.h to include the named header.  It is VERY important to
+# understand that this MUST be defined and the header MUST define internal
+# 'extern' functions with PNG_INTERNAL_FUNCTION.  This allows vendors to use
+# static library builds of libpng without clashing with the installed system
+# libpng; see the libpng-prefix option in configure.ac  The value of this
+# #define is included in pnglibconf.h
+#
+# NOTE: you must include the double quotes in the value and you must therefore
+# support the whole thing with single quotes.  If you need to use <> for a
+# header be aware that if you miss out the single quotes you will create a
+# random file.
+AM_CPPFLAGS += -DPNG_EXTENSION_HEADER='"%D%/neon.h"'
+endif

contrib/neon/neon.dfa

@@ -0,0 +1,41 @@
+# contrib/arm/neon.dfa
+#
+# This is a dfa fragment that is added to DFA_HARDWARE in Makefile.am and
+# consequently is inserted after scripts/pnglibconf.dfa to define extra,
+# hardware, options.
+#
+# The following are required in any fragment to record the copyright holder of
+# the *changes* and the appropriate email for bug reports.  Note that the
+# copyright here is just for the specific configuration; additional files may
+# have additional and/or different copyrights.
+#
+# Later settings (e.g. in user.dfa) will override these settings.
+#
+setting USER_PRIVATEBUILD default "John Bowler"
+setting USER_BUG_REPORTS default "jbowler@acm.org"
+#
+# These options are specific to the ARM NEON hardware optimizations.  At present
+# these optimizations depend on GCC specific pre-processing of an assembler (.S)
+# file so they probably won't work with other compilers.
+#
+# ARM_NEON_OPT: unset: check at compile time (__ARM_NEON__ must be defined by
+#                      the compiler, typically as a result of specifying
+#                      CC="gcc -mfpu=neon".)
+#                   0: disable (even if the CPU has a NEON FPU.)
+#                   1: check at run time (via ARM_NEON_{API,CHECK})
+#                   2: switch on unconditionally (inadvisable - instead pass
+#                      -mfpu=neon to GCC in CC)
+#           When building libpng avoid using any setting other than '0'; '1' is
+#           set automatically when either 'API' or 'CHECK' are configured in,
+#           '2' should not be necessary as -mfpu=neon will achieve the same
+#           effect as well as applying NEON optimizations to the rest of the
+#           libpng code.
+#           NOTE: any setting other than '0' requires ALIGNED_MEMORY
+# ARM_NEON_API:   (PNG_ARM_NEON == 1) allow the optimization to be switched on
+#                 with png_set_option
+# ARM_NEON_CHECK: (PNG_ARM_NEON == 1) compile a run-time check to see if Neon
+#                 extensions are supported. This is poorly supported and
+#                 deprecated - use the png_set_option API.
+setting ARM_NEON_OPT
+option ARM_NEON_API disabled enables SET_OPTION sets ARM_NEON_OPT 1
+option ARM_NEON_CHECK disabled sets ARM_NEON_OPT 1

contrib/neon/neon.h

@@ -0,0 +1,124 @@
+/* Compile time options.
+ * =====================
+ * In a multi-arch build the compiler may compile the code several times for the
+ * same object module, producing different binaries for different architectures.
+ * When this happens configure-time setting of the target host options cannot be
+ * done.  An example is iOS vs Android compilation of ARM NEON support.
+ *
+ * NOTE: symbol prefixing does not pass $(CFLAGS) to the preprocessor, because
+ * this is not possible with certain compilers (Oracle SUN OS CC), as a result
+ * it is necessary to ensure that all extern functions that *might* be used
+ * regardless of $(CFLAGS) get declared in this file.  The test on __ARM_NEON
+ * below is one example of this behavior because it is controlled by the
+ * presence or not of -mfpu=neon on the GCC command line, it is possible to do
+ * this in $(CC), e.g. "CC=gcc -mfpu=neon", but people who build libpng rarely
+ * do this.
+ */
+#ifndef PNG_ARM_NEON_OPT
+   /* ARM NEON optimizations are being controlled by the compiler settings,
+    * typically the target FPU.  If the FPU supports NEON instructions then the
+    * compiler will define __ARM_NEON and we can rely unconditionally on NEON
+    * instructions not crashing, otherwise we must disable use of NEON
+    * instructions.
+    *
+    * NOTE: at present these optimizations depend on 'ALIGNED_MEMORY', so they
+    * can only be turned on automatically if that is supported too.  If
+    * PNG_ARM_NEON_OPT is set in CPPFLAGS (to >0) then arm/arm_init.c will fail
+    * to compile with an appropriate #error if ALIGNED_MEMORY has been turned
+    * off.
+    *
+    * Note that older versions of GCC defined __ARM_NEON__; this is no longer
+    * supported.  Also 32-bit ARM versions of GCC required the NEON FPU mode to
+    * be turned on explicitly on the command line.  If this is not done (on
+    * 32-bit ARM) NEON code will not be included.
+    *
+    * To disable ARM_NEON optimizations entirely, and skip compiling the
+    * associated assembler code, pass --enable-arm-neon=no to configure
+    * or put -DPNG_ARM_NEON_OPT=0 in CPPFLAGS.
+    */
+#  if defined(__ARM_NEON) && defined(PNG_ALIGNED_MEMORY_SUPPORTED)
+#     define PNG_ARM_NEON_OPT 2
+#  else
+#     define PNG_ARM_NEON_OPT 0
+#  endif
+#endif
+
+#if PNG_ARM_NEON_OPT > 0
+   /* NEON optimizations are to be at least considered by libpng, so enable the
+    * callbacks to do this.
+    */
+#  define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_neon
+
+   /* By default the 'intrinsics' code in arm/filter_neon_intrinsics.c is used
+    * if possible - if __ARM_NEON is set and the compiler version is not known
+    * to be broken.  This is controlled by PNG_ARM_NEON_IMPLEMENTATION which can
+    * be:
+    *
+    *    1  The intrinsics code (the default with __ARM_NEON)
+    *    2  The hand coded assembler (the default without __ARM_NEON)
+    *
+    * It is possible to set PNG_ARM_NEON_IMPLEMENTATION in CPPFLAGS, however
+    * this is *NOT* supported and may cease to work even after a minor revision
+    * to libpng.  It *is* valid to do this for testing purposes, e.g. speed
+    * testing or a new compiler, but the results should be communicated to the
+    * libpng implementation list for incorporation in the next minor release.
+    */
+#  ifndef PNG_ARM_NEON_IMPLEMENTATION
+#     ifdef __ARM_NEON
+#        if defined(__clang__)
+            /* At present it is unknown by the libpng developers which versions
+             * of clang support the intrinsics, however some or perhaps all
+             * versions do not work with the assembler so this may be
+             * irrelevant, so just use the default (do nothing here.)
+             */
+#        elif defined(__GNUC__)
+            /* GCC 4.5.4 NEON support is known to be broken.  4.6.3 is known to
+             * work, so if this *is* GCC, or G++, look for a version >4.5
+             */
+#           if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 6)
+#              define PNG_ARM_NEON_IMPLEMENTATION 2
+#           endif /* no GNUC support */
+#        endif /* __GNUC__ */
+#     else /* !defined __ARM_NEON */
+         /* The 'intrinsics' code simply won't compile without compiler support
+          * and that support switches on __ARM_NEON, so use the assembler:
+          */
+#        define PNG_ARM_NEON_IMPLEMENTATION 2
+#     endif /* __ARM_NEON */
+#  endif /* !defined PNG_ARM_NEON_IMPLEMENTATION */
+
+#  ifndef PNG_ARM_NEON_IMPLEMENTATION
+      /* Use the intrinsics code by default. */
+#     define PNG_ARM_NEON_IMPLEMENTATION 1
+#  endif
+#endif /* PNG_ARM_NEON_OPT > 0 */
+
+#ifdef PNG_SET_OPTION_SUPPORTED
+#ifdef PNG_ARM_NEON_API_SUPPORTED
+#  define PNG_ARM_NEON   0 /* HARDWARE: ARM Neon SIMD instructions supported */
+#endif
+#endif /* SET_OPTION */
+
+#if PNG_ARM_NEON_OPT > 0
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_up_neon,
+   (png_alloc_size_t row_bytes, unsigned int bpp, png_bytep row,
+    png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_neon,
+   (png_alloc_size_t row_bytes, unsigned int bpp, png_bytep row,
+    png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_neon,
+   (png_alloc_size_t row_bytes, unsigned int bpp, png_bytep row,
+    png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_neon,
+   (png_alloc_size_t row_bytes, unsigned int bpp, png_bytep row,
+    png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_neon,
+   (png_alloc_size_t row_bytes, unsigned int bpp, png_bytep row,
+    png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_neon,
+   (png_alloc_size_t row_bytes, unsigned int bpp, png_bytep row,
+    png_const_bytep prev_row),PNG_EMPTY);
+PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_neon,
+   (png_alloc_size_t row_bytes, unsigned int bpp, png_bytep row,
+    png_const_bytep prev_row),PNG_EMPTY);
+#endif

contrib/pngminim/decoder/makefile

@@ -56,7 +56,7 @@ PNGSRCS=png$(C) pngerror$(C) pngget$(C) pngmem$(C) \
 	pngset$(C) pngtrans$(C)
 	
 # Standard headers
-PNGH   =png.h pngconf.h pngdebug.h pnginfo.h pngpriv.h pngstruct.h
+PNGH   =png.h pngconf.h pngdebug.h pnginfo.h pngpriv.h pngstruct.h pngchunk.h
 
 # Machine generated headers
 PNGCONF=pnglibconf.h

contrib/pngminim/decoder/pngusr.dfa

@@ -36,5 +36,7 @@ option FLOATING_ARITHMETIC off
 option SETJMP on
 option STDIO on
 option READ_EXPAND on
+option READ_IMAGE on
+option READ_DEINTERLACE on
 option READ_STRIP_16_TO_8 on
 option USER_LIMITS on

contrib/pngminim/decoder/pngusr.h

@@ -18,6 +18,7 @@
  * affect the API (so are not recorded in pnglibconf.h)
  */
 
+#define PNG_RELEASE_BUILD 1
 #define PNG_ALIGN_TYPE PNG_ALIGN_NONE
 
 #endif /* MINRDPNGCONF_H */

contrib/pngminim/encoder/makefile

@@ -56,7 +56,7 @@ PNGSRCS=png$(C) pngerror$(C) pngget$(C) pngmem$(C) \
 	pngwtran$(C) pngwutil$(C)
 
 # Standard headers
-PNGH   =png.h pngconf.h pngdebug.h pnginfo.h pngpriv.h pngstruct.h
+PNGH   =png.h pngconf.h pngdebug.h pnginfo.h pngpriv.h pngstruct.h pngchunk.h
 
 # Machine generated headers
 PNGCONF=pnglibconf.h

contrib/pngminim/encoder/pngusr.h

@@ -18,6 +18,7 @@
  * affect the API (so are not recorded in pnglibconf.h)
  */
 
+#define PNG_RELEASE_BUILD 1
 #define PNG_ALIGN_TYPE PNG_ALIGN_NONE
 
 #endif /* MINWRPNGCONF_H */

contrib/pngminim/preader/makefile

@@ -72,7 +72,7 @@ PNGSRCS=png$(C) pngerror$(C) pngget$(C) pngmem$(C) \
 	pngset$(C) pngtrans$(C)
 
 # Standard headers
-PNGH   =png.h pngconf.h pngdebug.h pnginfo.h pngpriv.h pngstruct.h
+PNGH   =png.h pngconf.h pngdebug.h pnginfo.h pngpriv.h pngstruct.h pngchunk.h
 
 # Machine generated headers
 PNGCONF=pnglibconf.h

contrib/pngminim/preader/pngusr.dfa

@@ -36,5 +36,6 @@ option STDIO on
 option READ_bKGD on
 option READ_GAMMA on
 option READ_EXPAND on
+option READ_DEINTERLACE on
 option READ_STRIP_16_TO_8 on
 option READ_GRAY_TO_RGB on

contrib/pngminim/preader/pngusr.h

@@ -18,6 +18,7 @@
  * affect the API (so are not recorded in pnglibconf.h)
  */
 
+#define PNG_RELEASE_BUILD 1
 #define PNG_ALIGN_TYPE PNG_ALIGN_NONE
 
 #endif /* MINPRDPNGCONF_H */

contrib/pngminus/README

@@ -146,7 +146,7 @@ The Turbo bug
 The end
 -------
 Willem van Schaik
-mailto:willem at schaik.com
+mailto:willem@schaik.com
 http://www.schaik.com/png/
 -------
 Oct 1999

contrib/pngminus/makefile.std

@@ -8,9 +8,9 @@ LD=$(CC)
 RM=rm -f
 
 #PNGPATH = /usr/local
-#PNGINC = -I$(PNGPATH)/include/libpng16
-#PNGLIB = -L$(PNGPATH)/lib -lpng16
-#PNGLIBS = $(PNGPATH)/lib/libpng16.a
+#PNGINC = -I$(PNGPATH)/include/libpng17
+#PNGLIB = -L$(PNGPATH)/lib -lpng17
+#PNGLIBS = $(PNGPATH)/lib/libpng17.a
 PNGINC = -I../..
 PNGLIB = -L../.. -lpng
 PNGLIBS = ../../libpng.a
@@ -23,8 +23,7 @@ ZINC = -I../../../zlib
 ZLIB = -L../../../zlib -lz
 ZLIBS = ../../../zlib/libz.a
 
-CPPFLAGS=$(PNGINC) $(ZINC)
-CFLAGS=
+CFLAGS=$(PNGINC) $(ZINC)
 LDLIBS=$(PNGLIB) $(ZLIB)
 LDLIBSS=$(PNGLIBS) $(ZLIBS)
 C=.c
@@ -38,7 +37,7 @@ E=
 all: png2pnm$(E) pnm2png$(E) png2pnm-static$(E) pnm2png-static$(E)
 
 png2pnm$(O): png2pnm$(C)
-	$(CC) -c $(CPPFLAGS) $(CFLAGS) png2pnm$(C)
+	$(CC) -c $(CFLAGS) png2pnm$(C)
 
 png2pnm$(E): png2pnm$(O)
 	$(LD) $(LDFLAGS) -o png2pnm$(E) png2pnm$(O) $(LDLIBS) -lm
@@ -47,7 +46,7 @@ png2pnm-static$(E): png2pnm$(O)
 	$(LD) $(LDFLAGS) -o png2pnm-static$(E) png2pnm$(O) $(LDLIBSS) -lm
 
 pnm2png$(O): pnm2png$(C)
-	$(CC) -c $(CPPFLAGS) $(CFLAGS) pnm2png$(C)
+	$(CC) -c $(CFLAGS) pnm2png$(C)
 
 pnm2png$(E): pnm2png$(O)
 	$(LD) $(LDFLAGS) -o pnm2png$(E) pnm2png$(O) $(LDLIBS) -lm

contrib/pngminus/makefile.tc3

@@ -7,8 +7,7 @@ LB=tlib
 RM=del
 CP=copy
 MODEL=l
-CPPFLAGS=-I..\libpng -I..\zlib
-CFLAGS=-O -m$(MODEL)
+CCFLAGS=-O -m$(MODEL) -I..\libpng -I..\zlib
 LDFLAGS=-m$(MODEL) -L..\libpng -L..\zlib
 C=.c
 O=.obj
@@ -20,13 +19,13 @@ E=.exe
 all: png2pnm$(E) pnm2png$(E)
 
 png2pnm$(O): png2pnm$(C)
-        $(CC) -c $(CPPFLAGS) $(CFLAGS) png2pnm$(C)
+        $(CC) -c $(CCFLAGS) png2pnm$(C)
 
 png2pnm$(E): png2pnm$(O)
         $(LD) $(LDFLAGS) png2pnm$(O) libpng$(L) zlib$(L)
 
 pnm2png$(O): pnm2png$(C)
-        $(CC) -c $(CPPFLAGS) $(CFLAGS) pnm2png$(C)
+        $(CC) -c $(CCFLAGS) pnm2png$(C)
 
 pnm2png$(E): pnm2png$(O)
         $(LD) $(LDFLAGS) pnm2png$(O) libpng$(L) zlib$(L)
@@ -36,3 +35,4 @@ clean:
         $(RM) *$(E)
 
 # End of makefile for png2pnm / pnm2png
+

contrib/pngminus/png2pnm.c

@@ -1,6 +1,6 @@
 /*
  *  png2pnm.c --- conversion from PNG-file to PGM/PPM-file
- *  copyright (C) 1999 by Willem van Schaik <willem at schaik.com>
+ *  copyright (C) 1999 by Willem van Schaik <willem@schaik.com>
  *
  *  version 1.0 - 1999.10.15 - First version.
  *
@@ -52,8 +52,7 @@
 
 int  main (int argc, char *argv[]);
 void usage ();
-BOOL png2pnm (FILE *png_file, FILE *pnm_file, FILE *alpha_file, BOOL raw,
-   BOOL alpha);
+BOOL png2pnm (FILE *png_file, FILE *pnm_file, FILE *alpha_file, BOOL raw, BOOL alpha);
 
 /*
  *  main
@@ -86,8 +85,7 @@ int main(int argc, char *argv[])
           if ((fp_al = fopen (argv[argi], "wb")) == NULL)
           {
             fprintf (stderr, "PNM2PNG\n");
-            fprintf (stderr, "Error:  can not create alpha-channel file %s\n",
-               argv[argi]);
+            fprintf (stderr, "Error:  can not create alpha-channel file %s\n", argv[argi]);
             exit (1);
           }
           break;
@@ -178,11 +176,9 @@ void usage()
   fprintf (stderr, "Usage:  png2pnm [options] <file>.png [<file>.pnm]\n");
   fprintf (stderr, "   or:  ... | png2pnm [options]\n");
   fprintf (stderr, "Options:\n");
-  fprintf (stderr,
-     "   -r[aw]   write pnm-file in binary format (P4/P5/P6) (default)\n");
+  fprintf (stderr, "   -r[aw]   write pnm-file in binary format (P4/P5/P6) (default)\n");
   fprintf (stderr, "   -n[oraw] write pnm-file in ascii format (P1/P2/P3)\n");
-  fprintf (stderr,
-     "   -a[lpha] <file>.pgm write PNG alpha channel as pgm-file\n");
+  fprintf (stderr, "   -a[lpha] <file>.pgm write PNG alpha channel as pgm-file\n");
   fprintf (stderr, "   -h | -?  print this help-information\n");
 }
 
@@ -190,8 +186,7 @@ void usage()
  *  png2pnm
  */
 
-BOOL png2pnm (FILE *png_file, FILE *pnm_file, FILE *alpha_file,
-    volatile BOOL raw, BOOL alpha)
+BOOL png2pnm (FILE *png_file, FILE *pnm_file, FILE *alpha_file, BOOL raw, BOOL alpha)
 {
   png_struct    *png_ptr = NULL;
   png_info        *info_ptr = NULL;
@@ -223,7 +218,7 @@ BOOL png2pnm (FILE *png_file, FILE *pnm_file, FILE *alpha_file,
 
   /* create png and info structures */
 
-  png_ptr = png_create_read_struct (png_get_libpng_ver(NULL),
+  png_ptr = png_create_read_struct (PNG_LIBPNG_VER_STRING,
     NULL, NULL, NULL);
   if (!png_ptr)
     return FALSE;   /* out of memory */
@@ -266,7 +261,7 @@ BOOL png2pnm (FILE *png_file, FILE *pnm_file, FILE *alpha_file,
     png_set_expand (png_ptr);
 
 #ifdef NJET
-  /* downgrade 16-bit images to 8-bit */
+  /* downgrade 16-bit images to 8 bit */
   if (bit_depth == 16)
     png_set_strip_16 (png_ptr);
   /* transform grayscale images into full-color */
@@ -320,18 +315,12 @@ BOOL png2pnm (FILE *png_file, FILE *pnm_file, FILE *alpha_file,
   /* row_bytes is the width x number of channels x (bit-depth / 8) */
   row_bytes = png_get_rowbytes (png_ptr, info_ptr);
 
-  if (height > ((size_t)(-1))/row_bytes) /* too big */ {
-    png_destroy_read_struct (&png_ptr, &info_ptr, NULL);
-    return FALSE;
-  }
-  if ((png_pixels = (png_byte *)
-     malloc (row_bytes * height * sizeof (png_byte))) == NULL) {
+  if ((png_pixels = (png_byte *) malloc (row_bytes * height * sizeof (png_byte))) == NULL) {
     png_destroy_read_struct (&png_ptr, &info_ptr, NULL);
     return FALSE;
   }
 
-  if ((row_pointers = (png_byte **)
-     malloc (height * sizeof (png_bytep))) == NULL)
+  if ((row_pointers = (png_byte **) malloc (height * sizeof (png_bytep))) == NULL)
   {
     png_destroy_read_struct (&png_ptr, &info_ptr, NULL);
     free (png_pixels);
@@ -340,7 +329,7 @@ BOOL png2pnm (FILE *png_file, FILE *pnm_file, FILE *alpha_file,
   }
 
   /* set the individual row_pointers to point at the correct offsets */
-  for (i = 0; i < ((int) height); i++)
+  for (i = 0; i < (height); i++)
     row_pointers[i] = png_pixels + i * row_bytes;
 
   /* now we can go ahead and just read the whole image */
@@ -383,9 +372,9 @@ BOOL png2pnm (FILE *png_file, FILE *pnm_file, FILE *alpha_file,
   /* write data to PNM file */
   pix_ptr = png_pixels;
 
-  for (row = 0; row < (int) height; row++)
+  for (row = 0; row < height; row++)
   {
-    for (col = 0; col < (int) width; col++)
+    for (col = 0; col < width; col++)
     {
       for (i = 0; i < (channels - alpha_present); i++)
       {

contrib/pngminus/pnm2png.c

@@ -1,6 +1,6 @@
 /*
  *  pnm2png.c --- conversion from PBM/PGM/PPM-file to PNG-file
- *  copyright (C) 1999 by Willem van Schaik <willem at schaik.com>
+ *  copyright (C) 1999 by Willem van Schaik <willem@schaik.com>
  *
  *  version 1.0 - 1999.10.15 - First version.
  *  version 1.1 - 2015.07.29 - Fixed leaks (Glenn Randers-Pehrson)
@@ -373,9 +373,6 @@ BOOL pnm2png (FILE *pnm_file, FILE *png_file, FILE *alpha_file, BOOL interlace,
     /* row_bytes is the width x number of channels x (bit-depth / 8) */
     row_bytes = width * channels * ((bit_depth <= 8) ? 1 : 2);
 
-  if (height > ((size_t)(-1))/row_bytes) /* too big */ {
-    return FALSE;
-  }
   if ((png_pixels = (png_byte *)
      malloc (row_bytes * height * sizeof (png_byte))) == NULL)
     return FALSE;

contrib/pngsuite/README

@@ -1,8 +1,8 @@
 
 pngsuite
 --------
-Copyright (c) Willem van Schaik, 1999, 2011, 2012
-Two images (ftbbn0g01.png and ftbbn0g02.png) are by Glenn Randers-Pehrson, 2012
+(c) Willem van Schaik, 1999, 2011, 2012
+Two images are by Glenn Randers-Pehrson, 2012
 
 Permission to use, copy, modify, and distribute these images for any
 purpose and without fee is hereby granted.
@@ -101,5 +101,5 @@ Testing basn6a16.png: PASS (1072 zero samples)
 libpng passes test
 
 Willem van Schaik
-<willem at schaik.com>
+<willem@schaik.com>
 October 1999

contrib/powerpc-vsx/README

@@ -1,81 +0,0 @@
-OPERATING SYSTEM SPECIFIC POWERPC DETECTION
---------------------------------------------
-
-Detection of the ability to execute POWERPC on processor requires
-operating system support.  (The information is not available in user mode.)
-
-Currently only this feature is supported only for linux platform.
-
-HOW TO USE THIS
----------------
-
-This directory contains C code fragments that can be included in powerpc/powerpc_init.c
-by setting the macro PNG_POWERPC_VSX_FILE to the file name in "" or <> at build
-time.  This setting is not recorded in pnglibconf.h and can be changed simply by
-rebuilding arm/arm_init.o with the required macro definition.
-
-For any of this code to be used the POWERPC code must be enabled and run time
-checks must be supported.  I.e.:
-
-#if PNG_POWERPC_VSX_OPT > 0
-#ifdef PNG_POWERPC_VSX_CHECK_SUPPORTED
-
-This is done in a 'configure' build by passing configure the argument:
-
-   --enable-powerpc-vsx=check
-
-FILE FORMAT
------------
-
-Each file documents its testing status as of the last time it was tested (which
-may have been a long time ago):
-
-STATUS: one of:
-   SUPPORTED: This indicates that the file is included in the regularly
-         performed test builds and bugs are fixed when discovered.
-   COMPILED: This indicates that the code did compile at least once.  See the
-         more detailed description for the extent to which the result was
-         successful.
-   TESTED: This means the code was fully compiled into the libpng test programs
-         and these were run at least once.
-
-BUG REPORTS: an email address to which to send reports of problems
-
-The file is a fragment of C code. It should not define any 'extern' symbols;
-everything should be static.  It must define the function:
-
-static int png_have_vsx(png_structp png_ptr);
-
-That function must return 1 if ARM NEON instructions are supported, 0 if not.
-It must not execute png_error unless it detects a bug.  A png_error will prevent
-the reading of the PNG and in the future, writing too.
-
-BUG REPORTS
------------
-
-If you mail a bug report for any file that is not SUPPORTED there may only be
-limited response.  Consider fixing it and sending a patch to fix the problem -
-this is more likely to result in action.
-
-CONTRIBUTIONS
--------------
-
-You may send contributions of new implementations to
-png-mng-implement@sourceforge.net.  Please write code in strict C90 C where
-possible.  Obviously OS dependencies are to be expected.  If you submit code you
-must have the authors permission and it must have a license that is acceptable
-to the current maintainer; in particular that license must permit modification
-and redistribution.
-
-Please try to make the contribution a single file and give the file a clear and
-unambiguous name that identifies the target OS.  If multiple files really are
-required put them all in a sub-directory.
-
-You must also be prepared to handle bug reports from users of the code, either
-by joining the png-mng-implement mailing list or by providing an email for the
-"BUG REPORTS" entry or both.  Please make sure that the header of the file
-contains the STATUS and BUG REPORTS fields as above.
-
-Please list the OS requirements as precisely as possible.  Ideally you should
-also list the environment in which the code has been tested and certainly list
-any environments where you suspect it might not work.

contrib/powerpc-vsx/linux.c

@@ -1,57 +0,0 @@
-/* contrib/powerpc-vsx/linux.c
- *
- * Copyright (c) 2017 Glenn Randers-Pehrson
- * Written by Vadim Barkov, 2017.
- * Last changed in libpng 1.6.29 [March 16, 2017]
- *
- * This code is released under the libpng license.
- * For conditions of distribution and use, see the disclaimer
- * and license in png.h
- *
- * STATUS: TESTED
- * BUG REPORTS: png-mng-implement@sourceforge.net
- *
- * png_have_vsx implemented for Linux by reading the widely available
- * pseudo-file /proc/cpuinfo. 
- *
- * This code is strict ANSI-C and is probably moderately portable; it does
- * however use <stdio.h> and it assumes that /proc/cpuinfo is never localized.
- */
-
-#include <stdio.h>
-#include <string.h>
-#include <stdlib.h>
-#include "png.h"
-
-#ifndef MAXLINE
-#  define MAXLINE 1024
-#endif
-
-static int
-png_have_vsx(png_structp png_ptr)
-{
-   FILE *f;
-
-   const char *string = "altivec supported";
-   char input[MAXLINE];
-   char *token = NULL;
-
-   PNG_UNUSED(png_ptr)
-
-   f = fopen("/proc/cpuinfo", "r");
-   if (f != NULL)
-   {
-      memset(input,0,MAXLINE);
-      while(fgets(input,MAXLINE,f) != NULL)
-      {
-         token = strstr(input,string);
-         if(token != NULL)
-            return 1;
-      }
-   }
-#ifdef PNG_WARNINGS_SUPPORTED
-   else
-      png_warning(png_ptr, "/proc/cpuinfo open failed");
-#endif
-   return 0;
-}

contrib/powerpc-vsx/linux_aux.c

@@ -1,36 +0,0 @@
-/* contrib/powerpc-vsx/linux_aux.c
- *
- * Copyright (c) 2017 Glenn Randers-Pehrson
- * Written by Vadim Barkov, 2017.
- * Last changed in libpng 1.6.29 [March 16, 2017]
- *
- * This code is released under the libpng license.
- * For conditions of distribution and use, see the disclaimer
- * and license in png.h
- *
- * STATUS: TESTED
- * BUG REPORTS: png-mng-implement@sourceforge.net
- *
- * png_have_vsx implemented for Linux by using the auxiliary vector mechanism.
- *
- * This code is strict ANSI-C and is probably moderately portable; it does
- * however use <stdio.h> and it assumes that /proc/cpuinfo is never localized.
- */
-
-#include "sys/auxv.h"
-#include "png.h"
-
-static int
-png_have_vsx(png_structp png_ptr)
-{
-
-   const unsigned long auxv = getauxval( AT_HWCAP );
-
-   PNG_UNUSED(png_ptr)
-
-   if(auxv & (PPC_FEATURE_HAS_ALTIVEC|PPC_FEATURE_HAS_VSX ))
-      return 1;
-   else
-      return 0;
-}
-

contrib/testpngs/makepngs.sh

@@ -15,10 +15,22 @@
 # ones that extend the code-coverage of libpng from the existing test files in
 # contrib/pngsuite.
 test -n "$MAKEPNG" || MAKEPNG=./makepng
+if test "$1" = "-v"
+then
+   verbose=1
+   shift
+else
+   verbose=
+fi
+what="$1"
+shift
+cmdline="$@"
 opts=
 
 mp(){
-   ${MAKEPNG} $opts $1 "$3" "$4" "$3-$4$2.png"
+   test -n "$verbose" &&
+      echo ${MAKEPNG} $opts $cmdline $1 "$3" "$4" "$3-$4$2.png"
+   ${MAKEPNG} $opts $cmdline $1 "$3" "$4" "$3-$4$2.png"
 }
 
 mpg(){
@@ -39,7 +51,7 @@ mptrans(){
    fi
 }
 
-case "$1" in
+case "$what" in
    --small)
       opts="--small";;&
 
@@ -72,19 +84,8 @@ case "$1" in
       done;;
 
    --coverage)
-      # Comments below indicate cases known to be required and not duplicated
-      # in other (required) cases; the aim is to get a minimal set that gives
-      # the maxium code coverage.
-      mpg none gray-alpha 8 # required: code coverage, sRGB opaque component
-      mpg none palette 8 # required: basic palette read
-      mpg 1.8 gray 2 # required: tests gamma threshold code
-      mpg 1.8 palette 2 # required: code coverage
-      mpg 1.8 palette 4 # required: code coverage
-      mpg 1.8 palette 8 # error limits only
-      mpg linear palette 8 # error limits only
-      mpg linear rgb-alpha 16 # error limits only
-      mpg sRGB palette 1 # required: code coverage
-      mpg sRGB rgb-alpha 16 # required: code coverage: pngread.c:2422 untested
+      # Extra images made to improve code coverage:
+      ${MAKEPNG} --insert sBIT 1 --tRNS gray 2 gray-2-sBIT-tRNS.png
       :;;
 
    *)

contrib/tools/README.txt

@@ -23,5 +23,4 @@ ORIGINAL AUTHORS
     of the people below claim any rights with regard to the contents of this
     directory.
 
-    John Bowler <jbowler at acm.org>
-    Glenn Randers-Pehrson <glennrp at users.sourceforge.net>
+    John Bowler <jbowler@acm.org>

contrib/tools/chkfmt

@@ -1,11 +1,4 @@
 #!/bin/sh
-
-# chkfmt
-#
-# COPYRIGHT: Written by John Cunningham Bowler, 2010.
-# To the extent possible under law, the author has waived all copyright and
-# related or neighboring rights to this work.  This work is published from:
-# United States.
 #
 # Check the format of the source files in the current directory - checks for a
 # line length of 80 characters max and no tab characters.

contrib/tools/chunkhash.c

@@ -0,0 +1,437 @@
+/* chunkhash.c -- build a perfect hash code for the chunk names on stdin.
+ *
+ * Last changed in libpng 1.7.0 [(PENDING RELEASE)]
+ *
+ * COPYRIGHT: Written by John Cunningham Bowler, 2014.
+ * To the extent possible under law, the author has waived all copyright and
+ * related or neighboring rights to this work.  This work is published from:
+ * United States.
+ *
+ * Feed this program all the known chunks that libpng must recognize.  It will
+ * generate an appropriate hash key for the macro in pngpriv.h  To generate the
+ * list of chunks currently defined in png.h do this:
+ *
+ *    sed -n -e 's/^#define png_\(....\) PNG_U32.*$/\1/p' png.h
+ *
+ * An alternative to using this program is to pipe the output of the above
+ * through gperf, however the code generated by perf is somewhat more verbose
+ * and it doesn't generate as good a result (however it's a heck of a lot faster
+ * than this program!)
+ */
+#include <stdio.h>
+#include <string.h>
+#include <stdlib.h>
+#include <png.h> /* for the libpng types for this platform */
+
+/* The machine generated file is checked in, so this works and it obtains the
+ * definition of the PNG_CHUNK_HASH macro.  To change this definition only ever
+ * alter the strings below!
+ */
+#include "chunkhash.h"
+
+/* The best_ variables are local variables defined in the functions below.  When
+ * used in libpng code, however, they are just constants.
+ */
+#define PNG_CHUNK_HASH_MASK best_mask
+#define PNG_CHUNK_HASH_C0 best_c[0]
+#define PNG_CHUNK_HASH_C1 best_c[1]
+#define PNG_CHUNK_HASH_C2 best_c[2]
+#define PNG_CHUNK_HASH_C3 best_c[3]
+
+/* These strings contain the C text to copy into the new version of
+ * contrib/tools/chunkhash.c
+ */
+static const char *strings[] = {
+"/* chunkhash.h -- a perfect hash code for the chunk names in png.h",
+" *",
+" * Last changed in libpng 1.7.0 [(PENDING RELEASE)]",
+" *",
+" * THIS IS A MACHINE GENERATED FILE.  See contrib/tools/chunkhash.c for",
+" * copyright and other information.",
+" *",
+" * USAGE:  To include the PNG_CHUNK_HASH macro and associated definitions:",
+" *",
+" * #define PNG_CHUNKHASH_DEFS",
+" * #include \"contrib/tools/chunkhash.h\"",
+" *",
+" * To define the png_chunk_hash array used by the macro:",
+" *",
+" * #define PNG_CHUNKHASH_CODE",
+" * #include \"contrib/tools/chunkhash.h\"",
+" *",
+" * One or both of the defines must be given except when building chunkhash",
+" * itself.",
+" */",
+"#ifdef PNG_CHUNKHASH_DEFS",
+"#ifndef PNG_CHUNKHASH_H",
+"#define PNG_CHUNKHASH_H",
+"/* A perfect hash code - returns a value 0..(PNG_KNOWN_CHUNK_COUNT-1) and is",
+" * generated by the ridiculously simple program in contrib/tools/chunkhash.c",
+" *",
+" * The hash code used here multiplies each byte by a different constant to",
+" * return a single number:",
+" *",
+" *    b0 * c[0] + b1 * [c1] + b2 * c[2] + b3 * c[3]",
+" *",
+" * The values of the constants are found by search using the a table of",
+" * primes, including 0, and may be (in fact are at present) 0 for some of the",
+" * bytes, the compiler is expected to optimize multiply by zero, or one!",
+" *",
+" * The lookup table reduces the sparse result of the hash calculation to the",
+" * correct index values.  The chunks are indexed in the string order of the",
+" * png_uint_32 chunk names.",
+" */",
+0, /* HEADER definitions go here */
+"",
+"extern const png_byte png_chunk_hash[PNG_CHUNK_HASH_MASK+1];",
+"#endif /* !PNG_CHUNKHASH_H */",
+"#endif /* PNG_CHUNKHASH_DEFS */",
+"",
+"#ifndef PNG_CHUNK_HASH",
+"#define PNG_CHUNK_HASH(chunk) (png_chunk_hash[PNG_CHUNK_HASH_MASK & (\\",
+"      ((chunk) >> 24) * PNG_CHUNK_HASH_C0 +\\",
+"      ((chunk) >> 16) * PNG_CHUNK_HASH_C1 +\\",
+"      ((chunk) >>  8) * PNG_CHUNK_HASH_C2 +\\",
+"      ((chunk)      ) * PNG_CHUNK_HASH_C3)])",
+"#endif",
+"",
+"#ifdef PNG_CHUNKHASH_CODE",
+"#ifndef PNG_CHUNKHASH_C",
+"#define PNG_CHUNKHASH_C",
+"const png_byte png_chunk_hash[PNG_CHUNK_HASH_MASK+1] = {",
+0, /* png.c definitions go here */
+"};",
+"#endif /* !PNG_CHUNKHASH_C */",
+"#endif /* PNG_CHUNKHASH_CODE */",
+0 /* end of file */
+};
+
+#define CHUNK_COUNT_MAX 32
+   /* If necessary this is easy to increase; just don't use a png_uint_32
+    * bitmask below, however it doesn't seem likely that the limit will be hit
+    * any time soon.
+    */
+
+static const png_byte
+chunk_hash[256] =
+{
+   0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21,
+   22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
+   41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59,
+   60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78,
+   79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97,
+   98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113,
+   114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128,
+   129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143,
+   144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158,
+   159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173,
+   174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188,
+   189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203,
+   204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218,
+   219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233,
+   234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248,
+   249, 250, 251, 252, 253, 254, 255
+};
+
+static void
+error(const char *string)
+{
+   fprintf(stderr, "chunkhash: %s\n", string);
+   exit(1);
+}
+
+/* Return a hash code for the given chunk; this is the same as the pngpriv.h
+ * macro, but parameterized.  The names of the parameters have to be chosen to
+ * match the above #defines.
+ */
+static png_uint_32
+test_hash(png_uint_32 chunk, const unsigned int *best_c, png_uint_32 best_mask)
+{
+#  define png_chunk_hash chunk_hash /* prevents lookup */
+      return PNG_CHUNK_HASH(chunk);
+#  undef png_chunk_hash
+}
+
+static void
+print_chunks(const png_uint_32 *chunks, unsigned int nchunks, png_uint_32 first,
+   png_uint_32 last, const unsigned int *best_c, png_uint_32 best_mask,
+   const png_byte *lut)
+{
+   /* 'last' is the last code to print plus 1, this is used to detect duplicates
+    * (which should not happen - if there are any it's an internal error!)
+    */
+   if (nchunks > 0)
+   {
+      /* Recursive algorithm; find the hash code of the next chunk, then print
+       * all remaining chunks with codes in the range first..code (including
+       * duplicates) and after print all remaining chunks with codes in the
+       * range (code+1)..last
+       */
+      const png_uint_32 chunk = *chunks++;
+#     define png_chunk_hash lut
+         const png_uint_32 code = PNG_CHUNK_HASH(chunk);
+#     undef png_chunk_hash
+
+      --nchunks;
+
+      /* The code might be out of the print range */
+      if (code >= first && code <= last)
+      {
+         if (code >= first) /* not time yet */
+            print_chunks(chunks, nchunks, first, code, best_c, best_mask, lut);
+
+         printf("#define PNG_CHUNK_%c%c%c%c_TAG %lu%s\n", (char)(chunk>>24),
+            (char)(chunk>>16), (char)(chunk>>8), (char)chunk,
+            (unsigned long)code, code == last ? " /* DUPLICATE */" : "");
+
+         if (code < last) /* still some to go */
+            print_chunks(chunks, nchunks, code+1, last, best_c, best_mask, lut);
+      }
+
+      else
+         print_chunks(chunks, nchunks, first, last, best_c, best_mask, lut);
+   }
+}
+
+static unsigned int primes[] =
+{
+   0, 1, 2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67,
+   71, 73, 79, 83, 89, 97, 101, 103, 107, 109, 113, 127, 131, 137, 139, 149,
+   151, 157, 163, 167, 173, 179, 181, 191, 193, 197, 199, 211, 223, 227, 229,
+   233, 239, 241, 251
+};
+
+#define NPRIMES ((sizeof primes)/(sizeof primes[0]))
+
+int
+main(void)
+{
+   /* Stdin should just contain chunk names, one per line */
+   png_uint_32 best_mask;
+   unsigned int best_c[4];
+   png_uint_32 chunks[CHUNK_COUNT_MAX];
+   png_byte known_chunk_count;
+
+   /* Not required; stop GCC whining: */
+   memset(best_c, 0xab, sizeof best_c);
+   best_mask = 0x12345678;
+
+   {
+      png_uint_32 this_chunk = 0;
+      png_byte n_chunks = 0;
+
+      for (;;)
+      {
+         int ch = getchar();
+
+         if ((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z'))
+         {
+            if (this_chunk > 0xffffffU)
+               error("chunk name too long");
+
+            this_chunk = (this_chunk << 8) + (unsigned)ch;
+         }
+
+         else
+         {
+            if (this_chunk > 0)
+            {
+               if (this_chunk <= 0xffffffU)
+                  error("chunk name too short");
+
+               if (n_chunks >= CHUNK_COUNT_MAX)
+                  error("too many chunks (check CHUNK_COUNT_MAX)");
+
+               chunks[n_chunks++] = this_chunk;
+               this_chunk = 0;
+            }
+         }
+
+         if (ch < 0)
+            break;
+      }
+
+      /* 22 is the number of chunks currently defined (excluding fRAc), at any
+       * time it should also be in PNG_KNOWN_CHUNK_COUNT, but this run of
+       * chunkhash may be trying to add a chunk so allow bigger numbers.
+       */
+#     ifdef PNG_KNOWN_CHUNK_COUNT
+#        define CHUNK_COUNT_MIN PNG_KNOWN_CHUNK_COUNT
+#     else
+#        define CHUNK_COUNT_MIN 26
+#     endif
+      if (n_chunks < CHUNK_COUNT_MIN)
+         error("too few chunks (expecting at least 26)");
+
+      known_chunk_count = n_chunks;
+   }
+
+   /* Exhaustive search of the hash parameters - in fact this isn't very slow at
+    * all.
+    */
+   {
+      unsigned int i1, c_zero = 0, c_one = 0;
+
+      for (i1=0; i1<NPRIMES; ++i1)
+      {
+         unsigned int i2;
+         unsigned int c[4];
+
+         fprintf(stderr, "TEST: %u\n", primes[i1]);
+         c[0] = primes[i1];
+
+         for (i2=0; i2<NPRIMES; ++i2)
+         {
+            unsigned int i3;
+
+            c[1] = primes[i2];
+
+            for (i3=0; i3<NPRIMES; ++i3)
+            {
+               unsigned int i4;
+
+               c[2] = primes[i3];
+
+               for (i4=0; i4<NPRIMES; ++i4)
+               {
+                  unsigned int i;
+                  png_uint_32 hash_mask = 0xffU;
+                  png_uint_32 hashes[CHUNK_COUNT_MAX];
+
+                  c[3] = primes[i4];
+
+                  while (hash_mask > 0xfU)
+                  {
+                     for (i=0; i<known_chunk_count; ++i)
+                     {
+                        unsigned int j;
+                        png_uint_32 hash = test_hash(chunks[i], c, hash_mask);
+
+                        for (j=0; j<i; ++j) if (hashes[j] == hash) goto next_i4;
+
+                        hashes[i] = hash;
+                     }
+
+                     hash_mask >>= 1;
+                  }
+
+               next_i4:
+                  if (hash_mask < 0xffU)
+                  {
+                     /* This worked */
+                     unsigned int best, c0 = 0, c1 = 0;
+
+                     hash_mask <<= 1;
+                     hash_mask += 1;
+
+                     for (i=0; i<4; ++i)
+                     {
+                        if (c[i] == 0)
+                           ++c0;
+
+                        else if (c[i] == 1)
+                           ++c1;
+                     }
+
+                     if (hash_mask == best_mask)
+                        best = (c0 > c_zero) || (c0 == c_zero && c1 > c_one);
+
+                     else
+                        best = (hash_mask < best_mask);
+
+                     if (best)
+                     {
+                        fprintf(stderr, "{%u,%u,%u,%u} & 0x%lx\n",
+                           c[0], c[1], c[2], c[3],
+                           (unsigned long)hash_mask);
+
+                        c_zero = c0;
+                        c_one = c1;
+                        best_mask = hash_mask;
+                        memcpy(best_c, c, sizeof best_c);
+                     }
+                  }
+               } /* i4 */
+            } /* i3 */
+         } /* i2 */
+      } /* i1 */
+   }
+
+   /* Calculate the LUT (png_chunk_hash) */
+   {
+      png_byte b;
+      png_byte lut[256];
+
+      /* A failure returns PNG_KNOWN_CHUNK_COUNT: */
+      memset(lut, known_chunk_count, sizeof lut);
+
+      for (b=0; b<known_chunk_count; ++b)
+         lut[test_hash(chunks[b], best_c, best_mask)] = b;
+
+      /* Validate the pngpriv.h hash function. */
+#     define png_chunk_hash lut
+         {
+            unsigned int i;
+
+            for (i=0; i<known_chunk_count; ++i)
+            {
+               png_uint_32 chunk = chunks[i];
+
+               if (PNG_CHUNK_HASH(chunk) != i)
+                  error("internal error: hash didn't work!");
+            }
+         }
+#     undef lut
+
+      /* Print all the results, first the stuff for pngpriv.h */
+      {
+         unsigned int i = 0;
+
+         while (strings[i] != 0)
+            puts(strings[i++]);
+
+         printf("#define PNG_CHUNK_HASH_MASK 0x%lxU\n",
+            (unsigned long)best_mask);
+         printf("#define PNG_CHUNK_HASH_C0 %u\n", best_c[0]);
+         printf("#define PNG_CHUNK_HASH_C1 %u\n", best_c[1]);
+         printf("#define PNG_CHUNK_HASH_C2 %u\n", best_c[2]);
+         printf("#define PNG_CHUNK_HASH_C3 %u\n", best_c[3]);
+
+         /* Print the hash codes of all the chunks */
+         putchar('\n');
+         print_chunks(chunks, known_chunk_count, 0, 0xffffffff, best_c,
+            best_mask, lut);
+         putchar('\n');
+         printf("#define PNG_KNOWN_CHUNK_COUNT %u\n", known_chunk_count);
+
+         while (strings[++i] != 0)
+            puts(strings[i]);
+
+         /* Now print the LUT */
+         fputs("   ", stdout);
+         {
+            unsigned int j;
+
+            for (j=0; j<=best_mask; ++j)
+            {
+               printf("%d", lut[j]);
+
+               if ((j % 16) == 15 && j < best_mask)
+                  printf(",\n   ");
+
+               else if (j < best_mask)
+                  printf(", ");
+
+               else
+                  printf("\n");
+            }
+         }
+
+         /* And the trailing text */
+         while (strings[++i] != 0)
+            puts(strings[i]);
+      }
+   }
+
+   exit(0);
+}

contrib/tools/chunkhash.h

@@ -0,0 +1,97 @@
+/* chunkhash.h -- a perfect hash code for the chunk names in png.h
+ *
+ * Last changed in libpng 1.7.0 [(PENDING RELEASE)]
+ *
+ * THIS IS A MACHINE GENERATED FILE.  See contrib/tools/chunkhash.c for
+ * copyright and other information.
+ *
+ * USAGE:  To include the PNG_CHUNK_HASH macro and associated definitions:
+ *
+ * #define PNG_CHUNKHASH_DEFS
+ * #include "contrib/tools/chunkhash.h"
+ *
+ * To define the png_chunk_hash array used by the macro:
+ *
+ * #define PNG_CHUNKHASH_CODE
+ * #include "contrib/tools/chunkhash.h"
+ *
+ * One or both of the defines must be given except when building chunkhash
+ * itself.
+ */
+#ifdef PNG_CHUNKHASH_DEFS
+#ifndef PNG_CHUNKHASH_H
+#define PNG_CHUNKHASH_H
+/* A perfect hash code - returns a value 0..(PNG_KNOWN_CHUNK_COUNT-1) and is
+ * generated by the ridiculously simple program in contrib/tools/chunkhash.c
+ *
+ * The hash code used here multiplies each byte by a different constant to
+ * return a single number:
+ *
+ *    b0 * c[0] + b1 * [c1] + b2 * c[2] + b3 * c[3]
+ *
+ * The values of the constants are found by search using the a table of
+ * primes, including 0, and may be (in fact are at present) 0 for some of the
+ * bytes, the compiler is expected to optimize multiply by zero, or one!
+ *
+ * The lookup table reduces the sparse result of the hash calculation to the
+ * correct index values.  The chunks are indexed in the string order of the
+ * png_uint_32 chunk names.
+ */
+#define PNG_CHUNK_HASH_MASK 0x3fU
+#define PNG_CHUNK_HASH_C0 103
+#define PNG_CHUNK_HASH_C1 1
+#define PNG_CHUNK_HASH_C2 0
+#define PNG_CHUNK_HASH_C3 1
+
+#define PNG_CHUNK_IDAT_TAG 0
+#define PNG_CHUNK_IEND_TAG 1
+#define PNG_CHUNK_IHDR_TAG 2
+#define PNG_CHUNK_PLTE_TAG 3
+#define PNG_CHUNK_bKGD_TAG 4
+#define PNG_CHUNK_cHRM_TAG 5
+#define PNG_CHUNK_fRAc_TAG 6
+#define PNG_CHUNK_gAMA_TAG 7
+#define PNG_CHUNK_gIFg_TAG 8
+#define PNG_CHUNK_gIFt_TAG 9
+#define PNG_CHUNK_gIFx_TAG 10
+#define PNG_CHUNK_hIST_TAG 11
+#define PNG_CHUNK_iCCP_TAG 12
+#define PNG_CHUNK_iTXt_TAG 13
+#define PNG_CHUNK_oFFs_TAG 14
+#define PNG_CHUNK_pCAL_TAG 15
+#define PNG_CHUNK_pHYs_TAG 16
+#define PNG_CHUNK_sBIT_TAG 17
+#define PNG_CHUNK_sCAL_TAG 18
+#define PNG_CHUNK_sPLT_TAG 19
+#define PNG_CHUNK_sRGB_TAG 20
+#define PNG_CHUNK_sTER_TAG 21
+#define PNG_CHUNK_tEXt_TAG 22
+#define PNG_CHUNK_tIME_TAG 23
+#define PNG_CHUNK_tRNS_TAG 24
+#define PNG_CHUNK_zTXt_TAG 25
+
+#define PNG_KNOWN_CHUNK_COUNT 26
+
+extern const png_byte png_chunk_hash[PNG_CHUNK_HASH_MASK+1];
+#endif /* !PNG_CHUNKHASH_H */
+#endif /* PNG_CHUNKHASH_DEFS */
+
+#ifndef PNG_CHUNK_HASH
+#define PNG_CHUNK_HASH(chunk) (png_chunk_hash[PNG_CHUNK_HASH_MASK & (\
+      ((chunk) >> 24) * PNG_CHUNK_HASH_C0 +\
+      ((chunk) >> 16) * PNG_CHUNK_HASH_C1 +\
+      ((chunk) >>  8) * PNG_CHUNK_HASH_C2 +\
+      ((chunk)      ) * PNG_CHUNK_HASH_C3)])
+#endif
+
+#ifdef PNG_CHUNKHASH_CODE
+#ifndef PNG_CHUNKHASH_C
+#define PNG_CHUNKHASH_C
+const png_byte png_chunk_hash[PNG_CHUNK_HASH_MASK+1] = {
+   26, 3, 26, 26, 26, 26, 26, 13, 26, 26, 26, 16, 26, 26, 26, 26,
+   26, 24, 12, 26, 18, 26, 26, 26, 26, 20, 26, 17, 26, 26, 25, 15,
+   26, 8, 14, 26, 26, 22, 26, 26, 1, 19, 5, 21, 26, 26, 9, 26,
+   26, 26, 10, 7, 26, 11, 26, 0, 26, 2, 23, 26, 26, 4, 26, 6
+};
+#endif /* !PNG_CHUNKHASH_C */
+#endif /* PNG_CHUNKHASH_CODE */

contrib/tools/dynamic-range.c

@@ -0,0 +1,208 @@
+/* dynamic-range.c
+ *
+ * Last changed in libpng 1.7.0
+ *
+ * COPYRIGHT: Written by John Cunningham Bowler, 2015
+ * To the extent possible under law, the author has waived all copyright and
+ * related or neighboring rights to this work.  This work is published from:
+ * United States.
+ *
+ * Find the dynamic range of a given gamma encoding given a (linear) precision
+ * and a maximum number of encoded values.
+ */
+#include <stdlib.h>
+#include <stdio.h>
+#include <float.h>
+#include <math.h>
+#include <assert.h>
+
+double range(unsigned int steps, double factor, double gamma)
+{
+   return pow((steps * (pow(factor, 1/gamma) - 1)), gamma);
+}
+
+double max_range_gamma(unsigned int steps, double factor, double *max_range,
+      double glo, double rlo, double gmid, double rmid, double ghi, double rhi)
+{
+   /* Given three values which contain a peak value (so rmid > rlo and rmid >
+    * rhi) find the peak by repeated division of the range.  The algorithm is to
+    * find the range for two gamma values mid-way between the two pairs
+    * (glo,gmid), (ghi,gmid) then find the max; this gives us a new glo/ghi
+    * which must be half the distance apart of the previous pair.
+    */
+   double gammas[5];
+   double ranges[5];
+
+   gammas[0] = glo;  ranges[0] = rlo;
+   gammas[2] = gmid; ranges[2] = rmid;
+   gammas[4] = ghi;  ranges[4] = rhi;
+
+   for (;;)
+   {
+      int i, m;
+
+      ranges[1] = range(steps, factor, gammas[1] = (gammas[0]+gammas[2])/2);
+      ranges[3] = range(steps, factor, gammas[3] = (gammas[2]+gammas[4])/2);
+
+      for (m=1, i=2; i<4; ++i)
+         if (ranges[i] >= ranges[m])
+            m = i;
+
+      assert(gammas[0] < gammas[m] && gammas[m] < gammas[4]);
+      assert(ranges[0] < ranges[m] && ranges[m] > ranges[4]);
+
+      gammas[0] = gammas[m-1]; ranges[0] = ranges[m-1];
+      gammas[4] = gammas[m+1]; ranges[4] = ranges[m+1];
+      gammas[2] = gammas[m];   ranges[2] = ranges[m];
+
+      if (((gammas[4] - gammas[0])/gammas[2]-1) < 3*DBL_EPSILON ||
+          ((ranges[2] - ranges[0])/ranges[2]-1) < 6*DBL_EPSILON)
+      {
+         *max_range = ranges[2];
+         return gammas[2];
+      }
+   }
+}
+
+double best_gamma(unsigned int values, double precision, double *best_range)
+{
+   /* The 'guess' gamma value is determined by the following formula, which is
+    * itself derived from linear regression using values returned by this
+    * program:
+    */
+   double gtry = values * precision / 2.736;
+   double rtry;
+
+   /* 'values' needs to be the number of steps after the first, we have to
+    * reserve the first value, 0, for 0, so subtract 2 from values.  precision
+    * must be adjusted to the step factor.
+    */
+   values -= 2U;
+   precision += 1;
+   rtry = range(values, precision, gtry);
+
+   /* Now find two values either side of gtry with a lower range. */
+   {
+      double glo, ghi, rlo, rhi, gbest, rbest;
+
+      glo = gtry;
+      do
+      {
+         glo *= 0.9;
+         rlo = range(values, precision, glo);
+      }
+      while (rlo >= rtry);
+
+      ghi = gtry;
+      do
+      {
+         ghi *= 1.1;
+         rhi = range(values, precision, ghi);
+      }
+      while (rhi >= rtry);
+
+      gbest = max_range_gamma(values, precision, &rbest,
+            glo, rlo, gtry, rtry, ghi, rhi);
+
+      *best_range = rbest / precision;
+      return gbest;
+   }
+}
+
+double linear_regression(double precision, double *bp)
+{
+   unsigned int values, count = 0;
+   double g_sum = 0, g2_sum = 0, v_sum = 0, v2_sum = 0, gv_sum = 0;
+
+   /* Perform simple linear regression to get:
+    *
+    *    gamma = a + b.values
+    */
+   for (values = 128; values < 65536; ++values, ++count)
+   {
+      double range;
+      double gamma = best_gamma(values, precision, &range);
+
+      g_sum += gamma;
+      g2_sum += gamma * gamma;
+      v_sum += values;
+      v2_sum += values * (double)values;
+      gv_sum += gamma * values;
+      /* printf("%u %g %g\n", values, gamma, range); */
+   }
+
+   g_sum /= count;
+   g2_sum /= count;
+   v_sum /= count;
+   v2_sum /= count;
+   gv_sum /= count;
+
+   {
+      double b = (gv_sum - g_sum * v_sum) / (v2_sum - v_sum * v_sum);
+      *bp = b;
+      return g_sum - b * v_sum;
+   }
+}
+
+int
+main(int argc, const char **argv)
+{
+   double precision = argc == 2 ? atof(argv[1]) : 0;
+
+   /* Perform a second linear regression here on b:
+    *
+    *    b = bA + bB * precision
+    */
+   if (precision == 0)
+   {
+      double b_sum = 0, b2_sum = 0, p_sum = 0, p2_sum = 0, bp_sum = 0,
+             a_sum = 0, count = 0;
+
+      for (precision = .001; precision <= 0.01; precision += .001, count += 1)
+      {
+         double b;
+         double a = linear_regression(precision, &b);
+
+         b_sum += b;
+         b2_sum += b * b;
+         p_sum += precision;
+         p2_sum += precision * precision;
+         bp_sum += b * precision;
+         a_sum += a;
+      }
+
+      b_sum /= count;
+      b2_sum /= count;
+      p_sum /= count;
+      p2_sum /= count;
+      bp_sum /= count;
+      a_sum /= count;
+
+      {
+         double bB = (bp_sum - b_sum * p_sum) / (p2_sum - p_sum * p_sum);
+         double bA = b_sum - bB * p_sum;
+
+         printf("a = %g, b = %g + precision/%g\n", a_sum, bA, 1/bB);
+      }
+   }
+
+   else
+   {
+      unsigned int bits;
+      double b;
+      double a = linear_regression(precision, &b);
+      printf("precision %g: gamma = %g + values*%g\n", precision, a, b);
+
+      /* For information, given a precision: */
+      for (bits=7U; bits <= 16U; ++bits)
+      {
+         unsigned int values = 1U<<bits;
+         double gamma = values*precision/2.736;
+         double r = range(values-2U, 1+precision, gamma);
+
+         printf("bits: %u, gamma: %g, range: 1:%g\n", bits, gamma, r);
+      }
+   }
+
+   return 0;
+}

contrib/tools/genpng.c

@@ -324,7 +324,7 @@ alpha_calc(const struct arg *arg, double x, double y)
                {
                   double wx = bicubic[abs(dx)][0];
 
-                  if (wx != 0 && arg->inside_fn(arg, x+dx/16, y+dy/16))
+                  if (wx != 0 && arg->inside_fn(arg, x+dx/16, y+dy/16)) 
                      alphay += wx;
                }
 
@@ -565,7 +565,7 @@ line_check(double x, double y, double x1, double y1, double x2, double y2,
    /* The dot product is the distance down the line, the cross product is
     * the distance away from the line:
     *
-    *    distance = |cross| / sqrt(len2)
+    *    distance = |cross| / sqrt(len2) 
     */
    cross = x * ly - y * lx;
 
@@ -722,7 +722,7 @@ pixel(png_uint_16p p, struct arg *args, int nargs, double x, double y)
    /* 'a' may be negative or greater than 1; if it is, negative clamp the
     * pixel to 0 if >1 clamp r/g/b:
     */
-   if (a > 0)
+   if (a > 0) 
    {
       if (a > 1)
       {

contrib/tools/png-fix-itxt.c

@@ -8,7 +8,7 @@
  * For conditions of distribution and use, see the disclaimer
  * and license in png.h
  *
- * Usage:
+ * Usage:            
  *
  *     png-fix-itxt.exe < bad.png > good.png
  *

contrib/tools/pngfix.c

@@ -1,8 +1,8 @@
 /* pngfix.c
  *
- * Copyright (c) 2014-2016 John Cunningham Bowler
+ * Copyright (c) 2014-2015 John Cunningham Bowler
  *
- * Last changed in libpng 1.6.26 [October 20, 2016]
+ * Last changed in libpng 1.6.18 [(PENDING RELEASE)]
  *
  * This code is released under the libpng license.
  * For conditions of distribution and use, see the disclaimer
@@ -137,7 +137,7 @@
 #define png_zTXt PNG_U32(122,  84,  88, 116)
 #endif
 
-/* The 8-byte signature as a pair of 32-bit quantities */
+/* The 8 byte signature as a pair of 32 bit quantities */
 #define sig1 PNG_U32(137,  80,  78,  71)
 #define sig2 PNG_U32( 13,  10,  26,  10)
 
@@ -159,7 +159,7 @@
  */
 #define UNREACHED 0
 
-/* 80-bit number handling - a PNG image can be up to (2^31-1)x(2^31-1) 8-byte
+/* 80-bit number handling - a PNG image can be up to (2^31-1)x(2^31-1) 8 byte
  * (16-bit RGBA) pixels in size; that's less than 2^65 bytes or 2^68 bits, so
  * arithmetic of 80-bit numbers is sufficient.  This representation uses an
  * arbitrary length array of png_uint_16 digits (0..65535).  The representation
@@ -319,13 +319,13 @@ uarb_mult32(uarb acc, int a_digits, uarb num, int n_digits, png_uint_32 val)
       a_digits = uarb_mult_digit(acc, a_digits, num, n_digits,
          (png_uint_16)(val & 0xffff));
 
+      /* Because n_digits and val are >0 the following must be true: */
+      assert(a_digits > 0);
+
       val >>= 16;
       if (val > 0)
          a_digits = uarb_mult_digit(acc+1, a_digits-1, num, n_digits,
             (png_uint_16)val) + 1;
-
-      /* Because n_digits and val are >0 the following must be true: */
-      assert(a_digits > 0);
    }
 
    return a_digits;
@@ -587,7 +587,7 @@ chunk_type_valid(png_uint_32 c)
    c &= ~PNG_U32(32,32,0,32);
    t = (c & ~0x1f1f1f1f) ^ 0x40404040;
 
-   /* Subtract 65 for each 8-bit quantity, this must not overflow
+   /* Subtract 65 for each 8 bit quantity, this must not overflow
     * and each byte must then be in the range 0-25.
     */
    c -= PNG_U32(65,65,65,65);
@@ -670,7 +670,7 @@ IDAT_list_extend(struct IDAT_list *tail)
 
       if (length < tail->length) /* arithmetic overflow */
          length = tail->length;
-
+            
       next = voidcast(IDAT_list*, malloc(IDAT_list_size(NULL, length)));
       CLEAR(*next);
 
@@ -924,7 +924,7 @@ emit_string(const char *str, FILE *out)
 
       else if (isspace(UCHAR_MAX & *str))
          putc('_', out);
-
+   
       else
          fprintf(out, "\\%.3o", *str);
 }
@@ -1824,7 +1824,7 @@ IDAT_init(struct IDAT * const idat, struct file * const file)
 }
 
 static png_uint_32
-rechunk_length(struct IDAT *idat, int start)
+rechunk_length(struct IDAT *idat)
    /* Return the length for the next IDAT chunk, taking into account
     * rechunking.
     */
@@ -1836,7 +1836,7 @@ rechunk_length(struct IDAT *idat, int start)
       const struct IDAT_list *cur;
       unsigned int count;
 
-      if (start)
+      if (idat->idat_index == 0) /* at the new chunk (first time) */
          return idat->idat_length; /* use the cache */
 
       /* Otherwise rechunk_length is called at the end of a chunk for the length
@@ -1948,7 +1948,7 @@ process_IDAT(struct file *file)
       list->count = 0;
       file->idat->idat_list_tail = list;
    }
-
+   
    /* And fill in the next IDAT information buffer. */
    list->lengths[(list->count)++] = file->chunk->chunk_length;
 
@@ -1995,7 +1995,7 @@ process_IDAT(struct file *file)
       idat->idat_index = 0; /* Index into chunk data */
 
       /* Update the chunk length to the correct value for the IDAT chunk: */
-      file->chunk->chunk_length = rechunk_length(idat, 1/*start*/);
+      file->chunk->chunk_length = rechunk_length(idat);
 
       /* Change the state to writing IDAT chunks */
       file->state = STATE_IDAT;
@@ -2221,7 +2221,7 @@ zlib_init(struct zlib *zlib, struct IDAT *idat, struct chunk *chunk,
    /* These values are sticky across reset (in addition to the stuff in the
     * first block, which is actually constant.)
     */
-   zlib->file_bits = 24;
+   zlib->file_bits = 16;
    zlib->ok_bits = 16; /* unset */
    zlib->cksum = 0; /* set when a checksum error is detected */
 
@@ -2304,12 +2304,10 @@ zlib_advance(struct zlib *zlib, png_uint_32 nbytes)
                zlib->file_bits = file_bits;
 
                /* Check against the existing value - it may not need to be
-                * changed.  Note that a bogus file_bits is allowed through once,
-                * to see if it works, but the window_bits value is set to 15,
-                * the maximum.
+                * changed.
                 */
                if (new_bits == 0) /* no change */
-                  zlib->window_bits = ((file_bits > 15) ? 15 : file_bits);
+                  zlib->window_bits = file_bits;
 
                else if (new_bits != file_bits) /* rewrite required */
                   bIn = (png_byte)((bIn & 0xf) + ((new_bits-8) << 4));
@@ -2330,7 +2328,8 @@ zlib_advance(struct zlib *zlib, png_uint_32 nbytes)
                if (bIn != b2)
                {
                   /* If the first byte wasn't changed this indicates an error in
-                   * the checksum calculation; signal this by setting 'cksum'.
+                   * the checksum calculation; signal this by setting file_bits
+                   * (not window_bits) to 0.
                    */
                   if (zlib->file_bits == zlib->window_bits)
                      zlib->cksum = 1;
@@ -2589,7 +2588,7 @@ zlib_run(struct zlib *zlib)
    {
       struct chunk *chunk = zlib->chunk;
       int rc;
-
+      
       assert(zlib->rewrite_offset < chunk->chunk_length);
 
       rc = zlib_advance(zlib, chunk->chunk_length - zlib->rewrite_offset);
@@ -3473,8 +3472,7 @@ read_callback(png_structp png_ptr, png_bytep buffer, size_t count)
                      /* Write another IDAT chunk.  Call rechunk_length to
                       * calculate the length required.
                       */
-                     length = chunk->chunk_length =
-                         rechunk_length(file->idat, 0/*end*/);
+                     length = chunk->chunk_length = rechunk_length(file->idat);
                      assert(type == png_IDAT);
                      file->write_count = 0; /* for the new chunk */
                      --(file->write_count); /* fake out the increment below */
@@ -4047,4 +4045,3 @@ main(void)
    return 77;
 }
 #endif /* PNG_SETJMP_SUPPORTED */
-

contrib/tools/reindent

@@ -1,25 +0,0 @@
-#!/bin/sh
-
-# reindent a libpng C source
-
-# COPYRIGHT: Written by Glenn Randers-Pehrson, 2016.
-# To the extent possible under law, the author has waived all copyright and
-# related or neighboring rights to this work.  This work is published from:
-# United States.
-
-# Usage:
-# reindent inputtabsize outputtabsize inputcontinuestring outputcontinuestring
-#
-# Assumes that continued lines begin with indentation plus one space, and
-# that continued comments begin with indentation plus " *".
-#
-# eg, to change libpng coding style from 3-space indentation with 4-space
-# continuations to 4-space indentation with 2-space continuations:
-#
-#  reindent 3 4 "\t " "  " < example.c > example.c_4_2
-# and to restore the file back to libpng coding style
-#  reindent 4 3 "  " "    " < example.c_4_2 > example.c_3_4
-
-unexpand --first-only --t $1 | \
-   sed -e "/^	*$3[^\*]/{s/$3/$4/}" | \
-   expand -t $2

contrib/tools/sRGB.h

@@ -14,7 +14,7 @@
  * All routines take and return a floating point value in the range
  * 0 to 1.0, doing a calculation according to the sRGB specification
  * (in fact the source of the numbers is the wikipedia article at
- * https://en.wikipedia.org/wiki/SRGB).
+ * http://en.wikipedia.org/wiki/SRGB).
  */
 static double
 sRGB_from_linear(double l)

contrib/tools/scale.c

@@ -0,0 +1,238 @@
+/* Given a target range and a source range work out an expression to scale from
+ * the source to the target of the form:
+ *
+ *    (number * mult + add)>>16
+ *
+ * The command arguments are:
+ *
+ *    scale target source
+ *
+ * and the program works out a pair of numbers, mult and add, that evaluate:
+ *
+ *          number * target
+ *   round( --------------- )
+ *              source
+ *
+ * exactly for number in the range 0..source
+ */
+#define _ISOC99_SOURCE 1
+
+#include <stdlib.h>
+#include <string.h>
+#include <stdio.h>
+#include <math.h>
+
+static double minerr;
+static unsigned long minmult, minadd, minshift;
+static long mindelta;
+
+static int
+test(unsigned long target, unsigned long source, unsigned long mult,
+   long add, unsigned long shift, long delta)
+{
+   unsigned long i;
+   double maxerr = 0;
+   double rs = (double)target/source;
+
+   for (i=0; i<=source; ++i)
+   {
+      unsigned long t = i*mult+add;
+      double err = fabs((t >> shift) - i*rs);
+
+      if (err > minerr)
+         return 0;
+
+      if (err > maxerr)
+         maxerr = err;
+   }
+
+   if (maxerr < minerr)
+   {
+      minerr = maxerr;
+      minmult = mult;
+      minadd = add;
+      minshift = shift;
+      mindelta = delta;
+   }
+
+   return maxerr < .5;
+}
+
+static int
+dotest(unsigned long target, unsigned long source, unsigned long mult,
+   long add, unsigned long shift, long delta, int print)
+{
+   if (test(target, source, mult, add, shift, delta))
+   {
+      if (print & 4)
+         printf("      {%11lu,%6ld /* >>%lu */ }, /* %lu/%lu */\n",
+            mult, add, shift, target, source);
+
+      else if (print & 2)
+         printf("      {%11lu,%6ld,%3lu }, /* %lu/%lu */\n",
+            mult, add, shift, target, source);
+
+      else if (print)
+         printf("number * %lu/%lu = (number * %lu + %ld) >> %lu [delta %ld]\n",
+            target, source, mult, add, shift, delta);
+
+      return 1;
+   }
+
+   return 0;
+}
+
+static int
+find(unsigned long target, unsigned long source, int print, int fixshift)
+{
+   unsigned long shift = 0;
+   unsigned long shiftlim = 0;
+
+   /* In the final math the sum is at most (source*mult+add) >> shift, so:
+    *
+    *    source*mult+add < 1<<32
+    *    mult < (1<<32)/source
+    *
+    * but:
+    *
+    *    mult = (target<<shift)/source
+    *
+    * so:
+    *
+    *    (target<<shift) < (1<<32)
+    */
+   if (fixshift < 0)
+      while ((target<<shiftlim) < 0x80000000U) ++shiftlim;
+
+   else
+      shift = shiftlim = (unsigned long)fixshift;
+
+   minerr = 1E8;
+
+   for (; shift<=shiftlim; ++shift)
+   {
+      unsigned long mult = ((target<<shift) + (source>>1)) / source;
+      long delta;
+      long limit = 1; /* seems to be sufficient */
+      long add, start, end;
+
+      end = 1<<shift;
+      start = -end;
+
+      for (add=start; add<=end; ++add)
+         if (dotest(target,source,mult,add,shift,0,print))
+            return 1;
+
+      for (delta=1; delta<=limit; ++delta)
+      {
+#        if 0
+            fprintf(stderr, "%lu/%lu: shift %lu, delta %lu\n", target, source,
+               shift, delta);
+#        endif
+
+         for (add=start; add<=end; ++add)
+         {
+            if (dotest(target, source, mult-delta, add, shift, -delta, print))
+               return 1;
+
+            if (dotest(target, source, mult+delta, add, shift, delta, print))
+               return 1;
+         }
+      }
+   }
+
+   if (print & 4)
+      printf("      {%11lu,%6ld /* >>%lu */ }, /* %lu/%lu ERROR: .5+%g*/\n",
+         minmult, minadd, minshift, target, source, minerr-.5);
+
+   else if (print & 2)
+      printf("      {%11lu,%6ld,%3lu }, /* %lu/%lu ERROR: .5+%g*/\n",
+         minmult, minadd, minshift, target, source, minerr-.5);
+
+   else if (print)
+      printf(
+         "number * %lu/%lu ~= (number * %lu + %ld) >> %lu +/-.5+%g [delta %ld]\n",
+         target, source, minmult, minadd, minshift, minerr-.5, mindelta);
+
+   return 0;
+}
+
+static void
+usage(const char *prog)
+{
+   fprintf(stderr,
+      "usage: %s {--denominator|--maxshift|--code} target {source}\n"
+      " For each 'source' prints 'mult' and 'add' such that:\n\n"
+      "   (number * mult + add) >> 16 = round(number*target/source)\n\n"
+      " for all integer values of number in the range 0..source.\n\n"
+      " --denominator: swap target and source (specify a single source first\n"
+      "                and follow with multiple targets.)\n"
+      "    --maxshift: find the lowest shift value that works for all the\n"
+      "                repeated 'source' values\n"
+      "        --code: output C code for array/structure initialization\n",
+      prog);
+   exit(1);
+}
+
+int
+main(int argc, const char **argv)
+{
+   int i, err = 0, maxshift = 0, firstsrc = 1, code = 0, denominator = 0;
+   unsigned long target, shift = 0;
+
+   while (argc > 1)
+   {
+      if (strcmp(argv[firstsrc], "--maxshift") == 0)
+      {
+         maxshift = 1;
+         ++firstsrc;
+      }
+
+      else if (strcmp(argv[firstsrc], "--code") == 0)
+      {
+         code = 1;
+         ++firstsrc;
+      }
+
+      else if (strcmp(argv[firstsrc], "--denominator") == 0)
+      {
+         denominator = 1;
+         ++firstsrc;
+      }
+
+      else
+         break;
+   }
+
+
+   if (argc < 2+firstsrc)
+      usage(argv[0]);
+
+   target = strtoul(argv[firstsrc++], 0, 0);
+   if (target == 0) usage(argv[0]);
+
+   for (i=firstsrc; i<argc; ++i)
+   {
+      unsigned long source = strtoul(argv[i], 0, 0);
+
+      if (source == 0) usage(argv[0]);
+
+      if (!find(denominator ? source : target, denominator ? target : source,
+         maxshift ? 0 : 1+code, -1))
+         err = 1;
+
+      if (minshift > shift) shift = minshift;
+   }
+
+   if (maxshift) for (i=firstsrc; i<argc; ++i)
+   {
+      unsigned long source = strtoul(argv[i], 0, 0);
+
+      if (!find(denominator ? source : target, denominator ? target : source,
+         code ? 4 : 1, shift))
+         err = 1;
+   }
+
+   /* Just an exit code - the printout above lists the problem */
+   return err;
+}

example.c

@@ -2,8 +2,8 @@
 #if 0 /* in case someone actually tries to compile this */
 
 /* example.c - an example of using libpng
- * Last changed in libpng 1.6.24 [August 4, 2016]
- * Maintained 1998-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.7.0 [(PENDING RELEASE)]
+ * Maintained 1998-2015 Glenn Randers-Pehrson
  * Maintained 1996, 1997 Andreas Dilger)
  * Written 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
  * To the extent possible under law, the authors have waived
@@ -39,7 +39,7 @@
 #include <string.h>
 #include <stdio.h>
 #include <png.h>
-#include <zlib.h>
+#include PNG_ZLIB_HEADER /* this is the header libpng was built with */
 
 int main(int argc, const char **argv)
 {
@@ -97,7 +97,7 @@ int main(int argc, const char **argv)
           */
          if (buffer != NULL &&
             png_image_finish_read(&image, NULL/*background*/, buffer,
-                0/*row_stride*/, NULL/*colormap*/) != 0)
+               0/*row_stride*/, NULL/*colormap*/) != 0)
          {
             /* Now write the image out to the second argument.  In the write
              * call 'convert_to_8bit' allows 16-bit data to be squashed down to
@@ -105,7 +105,7 @@ int main(int argc, const char **argv)
              * to the 8-bit format.
              */
             if (png_image_write_to_file(&image, argv[2], 0/*convert_to_8bit*/,
-                buffer, 0/*row_stride*/, NULL/*colormap*/) != 0)
+               buffer, 0/*row_stride*/, NULL/*colormap*/) != 0)
             {
                /* The image has been written successfully. */
                exit(0);
@@ -295,7 +295,7 @@ void read_png(FILE *fp, int sig_read)  /* File is already open */
     * was compiled with a compatible version of the library.  REQUIRED
     */
    png_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING,
-       png_voidp user_error_ptr, user_error_fn, user_warning_fn);
+      png_voidp user_error_ptr, user_error_fn, user_warning_fn);
 
    if (png_ptr == NULL)
    {
@@ -375,7 +375,7 @@ void read_png(FILE *fp, int sig_read)  /* File is already open */
     * low byte.
     */
 #ifdef PNG_READ_SCALE_16_TO_8_SUPPORTED
-   png_set_scale_16(png_ptr);
+    png_set_scale_16(png_ptr);
 #else
    png_set_strip_16(png_ptr);
 #endif
@@ -419,10 +419,10 @@ void read_png(FILE *fp, int sig_read)  /* File is already open */
 
    if (png_get_bKGD(png_ptr, info_ptr, &image_background) != 0)
       png_set_background(png_ptr, image_background,
-          PNG_BACKGROUND_GAMMA_FILE, 1, 1.0);
+                         PNG_BACKGROUND_GAMMA_FILE, 1, 1.0);
    else
       png_set_background(png_ptr, &my_background,
-          PNG_BACKGROUND_GAMMA_SCREEN, 0, 1.0);
+                         PNG_BACKGROUND_GAMMA_SCREEN, 0, 1.0);
 
    /* Some suggestions as to how to get a screen gamma value
     *
@@ -481,7 +481,7 @@ void read_png(FILE *fp, int sig_read)  /* File is already open */
          png_color std_color_cube[MAX_SCREEN_COLORS];
 
          png_set_quantize(png_ptr, std_color_cube, MAX_SCREEN_COLORS,
-             MAX_SCREEN_COLORS, NULL, 0);
+            MAX_SCREEN_COLORS, NULL, 0);
       }
       /* This reduces the image to the palette supplied in the file */
       else if (png_get_PLTE(png_ptr, info_ptr, &palette, &num_palette) != 0)
@@ -491,7 +491,7 @@ void read_png(FILE *fp, int sig_read)  /* File is already open */
          png_get_hIST(png_ptr, info_ptr, &histogram);
 
          png_set_quantize(png_ptr, palette, num_palette,
-             max_screen_colors, histogram, 0);
+                        max_screen_colors, histogram, 0);
       }
    }
 #endif /* READ_QUANTIZE */
@@ -522,17 +522,17 @@ void read_png(FILE *fp, int sig_read)  /* File is already open */
    png_set_swap(png_ptr);
 
    /* Add filler (or alpha) byte (before/after each RGB triplet) */
-   png_set_filler(png_ptr, 0xffff, PNG_FILLER_AFTER);
+   png_set_filler(png_ptr, 0xff, PNG_FILLER_AFTER);
 
-#ifdef PNG_READ_INTERLACING_SUPPORTED
+#ifdef PNG_READ_DEINTERLACE_SUPPORTED
    /* Turn on interlace handling.  REQUIRED if you are not using
     * png_read_image().  To see how to handle interlacing passes,
     * see the png_read_row() method below:
     */
    number_passes = png_set_interlace_handling(png_ptr);
-#else /* !READ_INTERLACING */
+#else
    number_passes = 1;
-#endif /* READ_INTERLACING */
+#endif /* READ_DEINTERLACE */
 
 
    /* Optional call to gamma correct and add the background to the palette
@@ -552,7 +552,7 @@ void read_png(FILE *fp, int sig_read)  /* File is already open */
 
    for (row = 0; row < height; row++)
       row_pointers[row] = png_malloc(png_ptr, png_get_rowbytes(png_ptr,
-          info_ptr));
+         info_ptr));
 
    /* Now it's time to read the image.  One of these methods is REQUIRED */
 #ifdef entire /* Read the entire image in one go */
@@ -574,10 +574,10 @@ void read_png(FILE *fp, int sig_read)  /* File is already open */
       {
 #ifdef sparkle /* Read the image using the "sparkle" effect. */
          png_read_rows(png_ptr, &row_pointers[y], NULL,
-             number_of_rows);
+            number_of_rows);
 #else no_sparkle /* Read the image using the "rectangle" effect */
          png_read_rows(png_ptr, NULL, &row_pointers[y],
-             number_of_rows);
+            number_of_rows);
 #endif no_sparkle /* Use only one of these two methods */
       }
 
@@ -614,7 +614,7 @@ initialize_png_reader(png_structp *png_ptr, png_infop *info_ptr)
     * linked libraries.
     */
    *png_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING,
-        png_voidp user_error_ptr, user_error_fn, user_warning_fn);
+       png_voidp user_error_ptr, user_error_fn, user_warning_fn);
 
    if (*png_ptr == NULL)
    {
@@ -649,14 +649,14 @@ initialize_png_reader(png_structp *png_ptr, png_infop *info_ptr)
     * the function png_get_progressive_ptr(png_ptr).
     */
    png_set_progressive_read_fn(*png_ptr, (void *)stream_data,
-       info_callback, row_callback, end_callback);
+      info_callback, row_callback, end_callback);
 
    return (OK);
 }
 
 int
 process_data(png_structp *png_ptr, png_infop *info_ptr,
-    png_bytep buffer, png_uint_32 length)
+   png_bytep buffer, png_uint_32 length)
 {
    if (setjmp(png_jmpbuf((*png_ptr))))
    {
@@ -691,7 +691,7 @@ info_callback(png_structp png_ptr, png_infop info)
 }
 
 row_callback(png_structp png_ptr, png_bytep new_row,
-    png_uint_32 row_num, int pass)
+   png_uint_32 row_num, int pass)
 {
    /*
     * This function is called for every row in the image.  If the
@@ -715,7 +715,7 @@ row_callback(png_structp png_ptr, png_bytep new_row,
     */
    png_bytep old_row = ((png_bytep *)our_data)[row_num];
 
-#ifdef PNG_READ_INTERLACING_SUPPORTED
+#ifdef PNG_READ_DEINTERLACE_SUPPORTED
    /* If both rows are allocated then copy the new row
     * data to the corresponding row data.
     */
@@ -744,7 +744,7 @@ row_callback(png_structp png_ptr, png_bytep new_row,
     * to pass the current row as new_row, and the function will combine
     * the old row and the new row.
     */
-#endif /* READ_INTERLACING */
+#endif /* READ_DEINTERLACE */
 }
 
 end_callback(png_structp png_ptr, png_infop info)
@@ -780,7 +780,7 @@ void write_png(char *file_name /* , ... other image information ... */)
     * in case we are using dynamically linked libraries.  REQUIRED.
     */
    png_ptr = png_create_write_struct(PNG_LIBPNG_VER_STRING,
-       png_voidp user_error_ptr, user_error_fn, user_warning_fn);
+      png_voidp user_error_ptr, user_error_fn, user_warning_fn);
 
    if (png_ptr == NULL)
    {
@@ -819,7 +819,7 @@ void write_png(char *file_name /* , ... other image information ... */)
     * png_init_io() here you would call
     */
    png_set_write_fn(png_ptr, (void *)user_io_ptr, user_write_fn,
-       user_IO_flush_function);
+      user_IO_flush_function);
    /* where user_io_ptr is a structure you want available to the callbacks */
 #endif no_streams /* Only use one initialization method */
 
@@ -842,7 +842,7 @@ void write_png(char *file_name /* , ... other image information ... */)
     * currently be PNG_COMPRESSION_TYPE_BASE and PNG_FILTER_TYPE_BASE. REQUIRED
     */
    png_set_IHDR(png_ptr, info_ptr, width, height, bit_depth, PNG_COLOR_TYPE_???,
-       PNG_INTERLACE_????, PNG_COMPRESSION_TYPE_BASE, PNG_FILTER_TYPE_BASE);
+      PNG_INTERLACE_????, PNG_COMPRESSION_TYPE_BASE, PNG_FILTER_TYPE_BASE);
 
    /* Set the palette if there is one.  REQUIRED for indexed-color images */
    palette = (png_colorp)png_malloc(png_ptr, PNG_MAX_PALETTE_LENGTH
@@ -983,21 +983,16 @@ void write_png(char *file_name /* , ... other image information ... */)
    png_uint_32 k, height, width;
 
    /* In this example, "image" is a one-dimensional array of bytes */
-
-   /* Guard against integer overflow */
-   if (height > PNG_SIZE_MAX/(width*bytes_per_pixel)) {
-      png_error(png_ptr, "Image_data buffer would be too large");
-   }
    png_byte image[height*width*bytes_per_pixel];
 
    png_bytep row_pointers[height];
 
    if (height > PNG_UINT_32_MAX/(sizeof (png_bytep)))
-      png_error (png_ptr, "Image is too tall to process in memory");
+     png_error (png_ptr, "Image is too tall to process in memory");
 
    /* Set up pointers into your "image" byte array */
    for (k = 0; k < height; k++)
-      row_pointers[k] = image + k*width*bytes_per_pixel;
+     row_pointers[k] = image + k*width*bytes_per_pixel;
 
    /* One of the following output methods is REQUIRED */
 

intel/filter_sse2_intrinsics.c

@@ -1,378 +0,0 @@
-
-/* filter_sse2_intrinsics.c - SSE2 optimized filter functions
- *
- * Copyright (c) 2016-2017 Glenn Randers-Pehrson
- * Written by Mike Klein and Matt Sarett
- * Derived from arm/filter_neon_intrinsics.c
- *
- * Last changed in libpng 1.6.29 [(PENDING RELEASE)]
- *
- * This code is released under the libpng license.
- * For conditions of distribution and use, see the disclaimer
- * and license in png.h
- */
-
-#include "../pngpriv.h"
-
-#ifdef PNG_READ_SUPPORTED
-
-#if PNG_INTEL_SSE_IMPLEMENTATION > 0
-
-#include <immintrin.h>
-
-/* Functions in this file look at most 3 pixels (a,b,c) to predict the 4th (d).
- * They're positioned like this:
- *    prev:  c b
- *    row:   a d
- * The Sub filter predicts d=a, Avg d=(a+b)/2, and Paeth predicts d to be
- * whichever of a, b, or c is closest to p=a+b-c.
- */
-
-static __m128i load4(const void* p) {
-   return _mm_cvtsi32_si128(*(const int*)p);
-}
-
-static void store4(void* p, __m128i v) {
-   *(int*)p = _mm_cvtsi128_si32(v);
-}
-
-static __m128i load3(const void* p) {
-   /* We'll load 2 bytes, then 1 byte,
-    * then mask them together, and finally load into SSE.
-    */
-   const png_uint_16* p01 = p;
-   const png_byte*    p2  = (const png_byte*)(p01+1);
-
-   png_uint_32 v012 = (png_uint_32)(*p01)
-                    | (png_uint_32)(*p2) << 16;
-   return load4(&v012);
-}
-
-static void store3(void* p, __m128i v) {
-   /* We'll pull from SSE as a 32-bit int, then write
-    * its bottom two bytes, then its third byte.
-    */
-   png_uint_32 v012;
-   store4(&v012, v);
-
-   png_uint_16* p01 = p;
-   png_byte*    p2  = (png_byte*)(p01+1);
-   *p01 = v012;
-   *p2  = v012 >> 16;
-}
-
-void png_read_filter_row_sub3_sse2(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev)
-{
-   /* The Sub filter predicts each pixel as the previous pixel, a.
-    * There is no pixel to the left of the first pixel.  It's encoded directly.
-    * That works with our main loop if we just say that left pixel was zero.
-    */
-   png_debug(1, "in png_read_filter_row_sub3_sse2");
-   __m128i a, d = _mm_setzero_si128();
-
-   int rb = row_info->rowbytes;
-   while (rb >= 4) {
-      a = d; d = load4(row);
-      d = _mm_add_epi8(d, a);
-      store3(row, d);
-
-      row += 3;
-      rb  -= 3;
-   }
-   if (rb > 0) {
-      a = d; d = load3(row);
-      d = _mm_add_epi8(d, a);
-      store3(row, d);
-
-      row += 3;
-      rb  -= 3;
-   }
-}
-
-void png_read_filter_row_sub4_sse2(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev)
-{
-   /* The Sub filter predicts each pixel as the previous pixel, a.
-    * There is no pixel to the left of the first pixel.  It's encoded directly.
-    * That works with our main loop if we just say that left pixel was zero.
-    */
-   png_debug(1, "in png_read_filter_row_sub4_sse2");
-   __m128i a, d = _mm_setzero_si128();
-
-   int rb = row_info->rowbytes;
-   while (rb > 0) {
-      a = d; d = load4(row);
-      d = _mm_add_epi8(d, a);
-      store4(row, d);
-
-      row += 4;
-      rb  -= 4;
-   }
-}
-
-void png_read_filter_row_avg3_sse2(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev)
-{
-   /* The Avg filter predicts each pixel as the (truncated) average of a and b.
-    * There's no pixel to the left of the first pixel.  Luckily, it's
-    * predicted to be half of the pixel above it.  So again, this works
-    * perfectly with our loop if we make sure a starts at zero.
-    */
-   png_debug(1, "in png_read_filter_row_avg3_sse2");
-   const __m128i zero = _mm_setzero_si128();
-   __m128i    b;
-   __m128i a, d = zero;
-
-   int rb = row_info->rowbytes;
-   while (rb >= 4) {
-             b = load4(prev);
-      a = d; d = load4(row );
-
-      /* PNG requires a truncating average, so we can't just use _mm_avg_epu8 */
-      __m128i avg = _mm_avg_epu8(a,b);
-      /* ...but we can fix it up by subtracting off 1 if it rounded up. */
-      avg = _mm_sub_epi8(avg, _mm_and_si128(_mm_xor_si128(a,b),
-                                            _mm_set1_epi8(1)));
-      d = _mm_add_epi8(d, avg);
-      store3(row, d);
-
-      prev += 3;
-      row  += 3;
-      rb   -= 3;
-   }
-   if (rb > 0) {
-             b = load3(prev);
-      a = d; d = load3(row );
-
-      /* PNG requires a truncating average, so we can't just use _mm_avg_epu8 */
-      __m128i avg = _mm_avg_epu8(a,b);
-      /* ...but we can fix it up by subtracting off 1 if it rounded up. */
-      avg = _mm_sub_epi8(avg, _mm_and_si128(_mm_xor_si128(a,b),
-                                            _mm_set1_epi8(1)));
-
-      d = _mm_add_epi8(d, avg);
-      store3(row, d);
-
-      prev += 3;
-      row  += 3;
-      rb   -= 3;
-   }
-}
-
-void png_read_filter_row_avg4_sse2(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev)
-{
-   /* The Avg filter predicts each pixel as the (truncated) average of a and b.
-    * There's no pixel to the left of the first pixel.  Luckily, it's
-    * predicted to be half of the pixel above it.  So again, this works
-    * perfectly with our loop if we make sure a starts at zero.
-    */
-   png_debug(1, "in png_read_filter_row_avg4_sse2");
-   const __m128i zero = _mm_setzero_si128();
-   __m128i    b;
-   __m128i a, d = zero;
-
-   int rb = row_info->rowbytes;
-   while (rb > 0) {
-             b = load4(prev);
-      a = d; d = load4(row );
-
-      /* PNG requires a truncating average, so we can't just use _mm_avg_epu8 */
-      __m128i avg = _mm_avg_epu8(a,b);
-      /* ...but we can fix it up by subtracting off 1 if it rounded up. */
-      avg = _mm_sub_epi8(avg, _mm_and_si128(_mm_xor_si128(a,b),
-                                            _mm_set1_epi8(1)));
-
-      d = _mm_add_epi8(d, avg);
-      store4(row, d);
-
-      prev += 4;
-      row  += 4;
-      rb   -= 4;
-   }
-}
-
-/* Returns |x| for 16-bit lanes. */
-static __m128i abs_i16(__m128i x) {
-#if PNG_INTEL_SSE_IMPLEMENTATION >= 2
-   return _mm_abs_epi16(x);
-#else
-   /* Read this all as, return x<0 ? -x : x.
-   * To negate two's complement, you flip all the bits then add 1.
-    */
-   __m128i is_negative = _mm_cmplt_epi16(x, _mm_setzero_si128());
-
-   /* Flip negative lanes. */
-   x = _mm_xor_si128(x, is_negative);
-
-   /* +1 to negative lanes, else +0. */
-   x = _mm_sub_epi16(x, is_negative);
-   return x;
-#endif
-}
-
-/* Bytewise c ? t : e. */
-static __m128i if_then_else(__m128i c, __m128i t, __m128i e) {
-#if PNG_INTEL_SSE_IMPLEMENTATION >= 3
-   return _mm_blendv_epi8(e,t,c);
-#else
-   return _mm_or_si128(_mm_and_si128(c, t), _mm_andnot_si128(c, e));
-#endif
-}
-
-void png_read_filter_row_paeth3_sse2(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev)
-{
-   /* Paeth tries to predict pixel d using the pixel to the left of it, a,
-    * and two pixels from the previous row, b and c:
-    *   prev: c b
-    *   row:  a d
-    * The Paeth function predicts d to be whichever of a, b, or c is nearest to
-    * p=a+b-c.
-    *
-    * The first pixel has no left context, and so uses an Up filter, p = b.
-    * This works naturally with our main loop's p = a+b-c if we force a and c
-    * to zero.
-    * Here we zero b and d, which become c and a respectively at the start of
-    * the loop.
-    */
-   png_debug(1, "in png_read_filter_row_paeth3_sse2");
-   const __m128i zero = _mm_setzero_si128();
-   __m128i c, b = zero,
-           a, d = zero;
-
-   int rb = row_info->rowbytes;
-   while (rb >= 4) {
-      /* It's easiest to do this math (particularly, deal with pc) with 16-bit
-       * intermediates.
-       */
-      c = b; b = _mm_unpacklo_epi8(load4(prev), zero);
-      a = d; d = _mm_unpacklo_epi8(load4(row ), zero);
-
-      /* (p-a) == (a+b-c - a) == (b-c) */
-      __m128i pa = _mm_sub_epi16(b,c);
-
-      /* (p-b) == (a+b-c - b) == (a-c) */
-      __m128i pb = _mm_sub_epi16(a,c);
-
-      /* (p-c) == (a+b-c - c) == (a+b-c-c) == (b-c)+(a-c) */
-      __m128i pc = _mm_add_epi16(pa,pb);
-
-      pa = abs_i16(pa);  /* |p-a| */
-      pb = abs_i16(pb);  /* |p-b| */
-      pc = abs_i16(pc);  /* |p-c| */
-
-      __m128i smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb));
-
-      /* Paeth breaks ties favoring a over b over c. */
-      __m128i nearest  = if_then_else(_mm_cmpeq_epi16(smallest, pa), a,
-                         if_then_else(_mm_cmpeq_epi16(smallest, pb), b,
-                                                                     c));
-
-      /* Note `_epi8`: we need addition to wrap modulo 255. */
-      d = _mm_add_epi8(d, nearest);
-      store3(row, _mm_packus_epi16(d,d));
-
-      prev += 3;
-      row  += 3;
-      rb   -= 3;
-   }
-   if (rb > 0) {
-      /* It's easiest to do this math (particularly, deal with pc) with 16-bit
-       * intermediates.
-       */
-      c = b; b = _mm_unpacklo_epi8(load3(prev), zero);
-      a = d; d = _mm_unpacklo_epi8(load3(row ), zero);
-
-      /* (p-a) == (a+b-c - a) == (b-c) */
-      __m128i pa = _mm_sub_epi16(b,c);
-
-      /* (p-b) == (a+b-c - b) == (a-c) */
-      __m128i pb = _mm_sub_epi16(a,c);
-
-      /* (p-c) == (a+b-c - c) == (a+b-c-c) == (b-c)+(a-c) */
-      __m128i pc = _mm_add_epi16(pa,pb);
-
-      pa = abs_i16(pa);  /* |p-a| */
-      pb = abs_i16(pb);  /* |p-b| */
-      pc = abs_i16(pc);  /* |p-c| */
-
-      __m128i smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb));
-
-      /* Paeth breaks ties favoring a over b over c. */
-      __m128i nearest  = if_then_else(_mm_cmpeq_epi16(smallest, pa), a,
-                         if_then_else(_mm_cmpeq_epi16(smallest, pb), b,
-                                                                     c));
-
-      /* Note `_epi8`: we need addition to wrap modulo 255. */
-      d = _mm_add_epi8(d, nearest);
-      store3(row, _mm_packus_epi16(d,d));
-
-      prev += 3;
-      row  += 3;
-      rb   -= 3;
-   }
-}
-
-void png_read_filter_row_paeth4_sse2(png_row_infop row_info, png_bytep row,
-   png_const_bytep prev)
-{
-   /* Paeth tries to predict pixel d using the pixel to the left of it, a,
-    * and two pixels from the previous row, b and c:
-    *   prev: c b
-    *   row:  a d
-    * The Paeth function predicts d to be whichever of a, b, or c is nearest to
-    * p=a+b-c.
-    *
-    * The first pixel has no left context, and so uses an Up filter, p = b.
-    * This works naturally with our main loop's p = a+b-c if we force a and c
-    * to zero.
-    * Here we zero b and d, which become c and a respectively at the start of
-    * the loop.
-    */
-   png_debug(1, "in png_read_filter_row_paeth4_sse2");
-   const __m128i zero = _mm_setzero_si128();
-   __m128i c, b = zero,
-           a, d = zero;
-
-   int rb = row_info->rowbytes;
-   while (rb > 0) {
-      /* It's easiest to do this math (particularly, deal with pc) with 16-bit
-       * intermediates.
-       */
-      c = b; b = _mm_unpacklo_epi8(load4(prev), zero);
-      a = d; d = _mm_unpacklo_epi8(load4(row ), zero);
-
-      /* (p-a) == (a+b-c - a) == (b-c) */
-      __m128i pa = _mm_sub_epi16(b,c);
-
-      /* (p-b) == (a+b-c - b) == (a-c) */
-      __m128i pb = _mm_sub_epi16(a,c);
-
-      /* (p-c) == (a+b-c - c) == (a+b-c-c) == (b-c)+(a-c) */
-      __m128i pc = _mm_add_epi16(pa,pb);
-
-      pa = abs_i16(pa);  /* |p-a| */
-      pb = abs_i16(pb);  /* |p-b| */
-      pc = abs_i16(pc);  /* |p-c| */
-
-      __m128i smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb));
-
-      /* Paeth breaks ties favoring a over b over c. */
-      __m128i nearest  = if_then_else(_mm_cmpeq_epi16(smallest, pa), a,
-                         if_then_else(_mm_cmpeq_epi16(smallest, pb), b,
-                                                                     c));
-
-      /* Note `_epi8`: we need addition to wrap modulo 255. */
-      d = _mm_add_epi8(d, nearest);
-      store4(row, _mm_packus_epi16(d,d));
-
-      prev += 4;
-      row  += 4;
-      rb   -= 4;
-   }
-}
-
-#endif /* PNG_INTEL_SSE_IMPLEMENTATION > 0 */
-#endif /* READ */

intel/intel_init.c

@@ -1,53 +0,0 @@
-
-/* intel_init.c - SSE2 optimized filter functions
- *
- * Copyright (c) 2016-2017 Glenn Randers-Pehrson
- * Written by Mike Klein and Matt Sarett, Google, Inc.
- * Derived from arm/arm_init.c
- *
- * Last changed in libpng 1.6.29 [(PENDING RELEASE)]
- *
- * This code is released under the libpng license.
- * For conditions of distribution and use, see the disclaimer
- * and license in png.h
- */
-
-#include "../pngpriv.h"
-
-#ifdef PNG_READ_SUPPORTED
-#if PNG_INTEL_SSE_IMPLEMENTATION > 0
-
-void
-png_init_filter_functions_sse2(png_structp pp, unsigned int bpp)
-{
-   /* The techniques used to implement each of these filters in SSE operate on
-    * one pixel at a time.
-    * So they generally speed up 3bpp images about 3x, 4bpp images about 4x.
-    * They can scale up to 6 and 8 bpp images and down to 2 bpp images,
-    * but they'd not likely have any benefit for 1bpp images.
-    * Most of these can be implemented using only MMX and 64-bit registers,
-    * but they end up a bit slower than using the equally-ubiquitous SSE2.
-   */
-   png_debug(1, "in png_init_filter_functions_sse2");
-   if (bpp == 3)
-   {
-      pp->read_filter[PNG_FILTER_VALUE_SUB-1] = png_read_filter_row_sub3_sse2;
-      pp->read_filter[PNG_FILTER_VALUE_AVG-1] = png_read_filter_row_avg3_sse2;
-      pp->read_filter[PNG_FILTER_VALUE_PAETH-1] =
-         png_read_filter_row_paeth3_sse2;
-   }
-   else if (bpp == 4)
-   {
-      pp->read_filter[PNG_FILTER_VALUE_SUB-1] = png_read_filter_row_sub4_sse2;
-      pp->read_filter[PNG_FILTER_VALUE_AVG-1] = png_read_filter_row_avg4_sse2;
-      pp->read_filter[PNG_FILTER_VALUE_PAETH-1] =
-          png_read_filter_row_paeth4_sse2;
-   }
-
-   /* No need optimize PNG_FILTER_VALUE_UP.  The compiler should
-    * autovectorize.
-    */
-}
-
-#endif /* PNG_INTEL_SSE_IMPLEMENTATION > 0 */
-#endif /* PNG_READ_SUPPORTED */

libpngpf.3

@@ -1,6 +1,6 @@
-.TH LIBPNGPF 3 "April 22, 2017"
+.TH LIBPNGPF 3 "November 12, 2015"
 .SH NAME
-libpng \- Portable Network Graphics (PNG) Reference Library 1.6.30beta02
+libpng \- Portable Network Graphics (PNG) Reference Library 1.7.0beta68
 (private functions)
 .SH SYNOPSIS
 \fB#include \fI"pngpriv.h"

mips/filter_msa_intrinsics.c

@@ -1,807 +0,0 @@
-
-/* filter_msa_intrinsics.c - MSA optimised filter functions
- *
- * Copyright (c) 2016 Glenn Randers-Pehrson
- * Written by Mandar Sahastrabuddhe, August 2016.
- * Last changed in libpng 1.6.25 [September 1, 2016]
- *
- * This code is released under the libpng license.
- * For conditions of distribution and use, see the disclaimer
- * and license in png.h
- */
-#include <stdio.h>
-#include <stdint.h>
-#include "../pngpriv.h"
-
-#ifdef PNG_READ_SUPPORTED
-
-/* This code requires -mfpu=msa on the command line: */
-#if PNG_MIPS_MSA_IMPLEMENTATION == 1 /* intrinsics code from pngpriv.h */
-
-#include <msa.h>
-
-/* libpng row pointers are not necessarily aligned to any particular boundary,
- * however this code will only work with appropriate alignment. mips/mips_init.c
- * checks for this (and will not compile unless it is done). This code uses
- * variants of png_aligncast to avoid compiler warnings.
- */
-#define png_ptr(type,pointer) png_aligncast(type *,pointer)
-#define png_ptrc(type,pointer) png_aligncastconst(const type *,pointer)
-
-/* The following relies on a variable 'temp_pointer' being declared with type
- * 'type'.  This is written this way just to hide the GCC strict aliasing
- * warning; note that the code is safe because there never is an alias between
- * the input and output pointers.
- */
-#define png_ldr(type,pointer)\
-   (temp_pointer = png_ptr(type,pointer), *temp_pointer)
-
-#if PNG_MIPS_MSA_OPT > 0
-
-#ifdef CLANG_BUILD
-   #define MSA_SRLI_B(a, b)   __msa_srli_b((v16i8) a, b)
-
-   #define LW(psrc)                              \
-   ( {                                           \
-       uint8_t *psrc_lw_m = (uint8_t *) (psrc);  \
-       uint32_t val_m;                           \
-                                                 \
-       asm volatile (                            \
-           "lw  %[val_m],  %[psrc_lw_m]  \n\t"   \
-                                                 \
-           : [val_m] "=r" (val_m)                \
-           : [psrc_lw_m] "m" (*psrc_lw_m)        \
-       );                                        \
-                                                 \
-       val_m;                                    \
-   } )
-
-   #define SH(val, pdst)                         \
-   {                                             \
-       uint8_t *pdst_sh_m = (uint8_t *) (pdst);  \
-       uint16_t val_m = (val);                   \
-                                                 \
-       asm volatile (                            \
-           "sh  %[val_m],  %[pdst_sh_m]  \n\t"   \
-                                                 \
-           : [pdst_sh_m] "=m" (*pdst_sh_m)       \
-           : [val_m] "r" (val_m)                 \
-       );                                        \
-   }
-
-   #define SW(val, pdst)                         \
-   {                                             \
-       uint8_t *pdst_sw_m = (uint8_t *) (pdst);  \
-       uint32_t val_m = (val);                   \
-                                                 \
-       asm volatile (                            \
-           "sw  %[val_m],  %[pdst_sw_m]  \n\t"   \
-                                                 \
-           : [pdst_sw_m] "=m" (*pdst_sw_m)       \
-           : [val_m] "r" (val_m)                 \
-       );                                        \
-   }
-
-       #if (__mips == 64)
-        #define SD(val, pdst)                         \
-        {                                             \
-            uint8_t *pdst_sd_m = (uint8_t *) (pdst);  \
-            uint64_t val_m = (val);                   \
-                                                      \
-            asm volatile (                            \
-                "sd  %[val_m],  %[pdst_sd_m]  \n\t"   \
-                                                      \
-                : [pdst_sd_m] "=m" (*pdst_sd_m)       \
-                : [val_m] "r" (val_m)                 \
-            );                                        \
-        }
-    #else
-        #define SD(val, pdst)                                          \
-        {                                                              \
-            uint8_t *pdst_sd_m = (uint8_t *) (pdst);                   \
-            uint32_t val0_m, val1_m;                                   \
-                                                                       \
-            val0_m = (uint32_t) ((val) & 0x00000000FFFFFFFF);          \
-            val1_m = (uint32_t) (((val) >> 32) & 0x00000000FFFFFFFF);  \
-                                                                       \
-            SW(val0_m, pdst_sd_m);                                     \
-            SW(val1_m, pdst_sd_m + 4);                                 \
-        }
-    #endif
-#else
-   #define MSA_SRLI_B(a, b)   (a >> b)
-
-#if (__mips_isa_rev >= 6)
-   #define LW(psrc)                              \
-   ( {                                           \
-       uint8_t *psrc_lw_m = (uint8_t *) (psrc);  \
-       uint32_t val_m;                           \
-                                                 \
-       asm volatile (                            \
-           "lw  %[val_m],  %[psrc_lw_m]  \n\t"   \
-                                                 \
-           : [val_m] "=r" (val_m)                \
-           : [psrc_lw_m] "m" (*psrc_lw_m)        \
-       );                                        \
-                                                 \
-       val_m;                                    \
-   } )
-
-   #define SH(val, pdst)                         \
-   {                                             \
-       uint8_t *pdst_sh_m = (uint8_t *) (pdst);  \
-       uint16_t val_m = (val);                   \
-                                                 \
-       asm volatile (                            \
-           "sh  %[val_m],  %[pdst_sh_m]  \n\t"   \
-                                                 \
-           : [pdst_sh_m] "=m" (*pdst_sh_m)       \
-           : [val_m] "r" (val_m)                 \
-       );                                        \
-   }
-
-   #define SW(val, pdst)                         \
-   {                                             \
-       uint8_t *pdst_sw_m = (uint8_t *) (pdst);  \
-       uint32_t val_m = (val);                   \
-                                                 \
-       asm volatile (                            \
-           "sw  %[val_m],  %[pdst_sw_m]  \n\t"   \
-                                                 \
-           : [pdst_sw_m] "=m" (*pdst_sw_m)       \
-           : [val_m] "r" (val_m)                 \
-       );                                        \
-   }
-
-   #if (__mips == 64)
-        #define SD(val, pdst)                         \
-        {                                             \
-            uint8_t *pdst_sd_m = (uint8_t *) (pdst);  \
-            uint64_t val_m = (val);                   \
-                                                      \
-            asm volatile (                            \
-                "sd  %[val_m],  %[pdst_sd_m]  \n\t"   \
-                                                      \
-                : [pdst_sd_m] "=m" (*pdst_sd_m)       \
-                : [val_m] "r" (val_m)                 \
-            );                                        \
-        }
-    #else
-        #define SD(val, pdst)                                          \
-        {                                                              \
-            uint8_t *pdst_sd_m = (uint8_t *) (pdst);                   \
-            uint32_t val0_m, val1_m;                                   \
-                                                                       \
-            val0_m = (uint32_t) ((val) & 0x00000000FFFFFFFF);          \
-            val1_m = (uint32_t) (((val) >> 32) & 0x00000000FFFFFFFF);  \
-                                                                       \
-            SW(val0_m, pdst_sd_m);                                     \
-            SW(val1_m, pdst_sd_m + 4);                                 \
-        }
-    #endif
-#else  // !(__mips_isa_rev >= 6)
-   #define LW(psrc)                              \
-   ( {                                           \
-       uint8_t *psrc_lw_m = (uint8_t *) (psrc);  \
-       uint32_t val_m;                           \
-                                                 \
-       asm volatile (                            \
-           "ulw  %[val_m],  %[psrc_lw_m]  \n\t"  \
-                                                 \
-           : [val_m] "=r" (val_m)                \
-           : [psrc_lw_m] "m" (*psrc_lw_m)        \
-       );                                        \
-                                                 \
-       val_m;                                    \
-   } )
-
-   #define SH(val, pdst)                         \
-   {                                             \
-       uint8_t *pdst_sh_m = (uint8_t *) (pdst);  \
-       uint16_t val_m = (val);                   \
-                                                 \
-       asm volatile (                            \
-           "ush  %[val_m],  %[pdst_sh_m]  \n\t"  \
-                                                 \
-           : [pdst_sh_m] "=m" (*pdst_sh_m)       \
-           : [val_m] "r" (val_m)                 \
-       );                                        \
-   }
-
-   #define SW(val, pdst)                         \
-   {                                             \
-       uint8_t *pdst_sw_m = (uint8_t *) (pdst);  \
-       uint32_t val_m = (val);                   \
-                                                 \
-       asm volatile (                            \
-           "usw  %[val_m],  %[pdst_sw_m]  \n\t"  \
-                                                 \
-           : [pdst_sw_m] "=m" (*pdst_sw_m)       \
-           : [val_m] "r" (val_m)                 \
-       );                                        \
-   }
-
-   #define SD(val, pdst)                                          \
-    {                                                              \
-        uint8_t *pdst_sd_m = (uint8_t *) (pdst);                   \
-        uint32_t val0_m, val1_m;                                   \
-                                                                   \
-        val0_m = (uint32_t) ((val) & 0x00000000FFFFFFFF);          \
-        val1_m = (uint32_t) (((val) >> 32) & 0x00000000FFFFFFFF);  \
-                                                                   \
-        SW(val0_m, pdst_sd_m);                                     \
-        SW(val1_m, pdst_sd_m + 4);                                 \
-    }
-
-    #define SW_ZERO(pdst)                      \
-    {                                          \
-        uint8_t *pdst_m = (uint8_t *) (pdst);  \
-                                               \
-        asm volatile (                         \
-            "usw  $0,  %[pdst_m]  \n\t"        \
-                                               \
-            : [pdst_m] "=m" (*pdst_m)          \
-            :                                  \
-        );                                     \
-    }
-#endif  // (__mips_isa_rev >= 6)
-#endif
-
-#define LD_B(RTYPE, psrc) *((RTYPE *) (psrc))
-#define LD_UB(...) LD_B(v16u8, __VA_ARGS__)
-#define LD_B2(RTYPE, psrc, stride, out0, out1)  \
-{                                               \
-    out0 = LD_B(RTYPE, (psrc));                 \
-    out1 = LD_B(RTYPE, (psrc) + stride);        \
-}
-#define LD_UB2(...) LD_B2(v16u8, __VA_ARGS__)
-#define LD_B4(RTYPE, psrc, stride, out0, out1, out2, out3)   \
-{                                                            \
-    LD_B2(RTYPE, (psrc), stride, out0, out1);                \
-    LD_B2(RTYPE, (psrc) + 2 * stride , stride, out2, out3);  \
-}
-#define LD_UB4(...) LD_B4(v16u8, __VA_ARGS__)
-
-#define ST_B(RTYPE, in, pdst) *((RTYPE *) (pdst)) = (in)
-#define ST_UB(...) ST_B(v16u8, __VA_ARGS__)
-#define ST_B2(RTYPE, in0, in1, pdst, stride)  \
-{                                             \
-    ST_B(RTYPE, in0, (pdst));                 \
-    ST_B(RTYPE, in1, (pdst) + stride);        \
-}
-#define ST_UB2(...) ST_B2(v16u8, __VA_ARGS__)
-#define ST_B4(RTYPE, in0, in1, in2, in3, pdst, stride)    \
-{                                                         \
-    ST_B2(RTYPE, in0, in1, (pdst), stride);               \
-    ST_B2(RTYPE, in2, in3, (pdst) + 2 * stride, stride);  \
-}
-#define ST_UB4(...) ST_B4(v16u8, __VA_ARGS__)
-
-#define ADD2(in0, in1, in2, in3, out0, out1)  \
-{                                             \
-    out0 = in0 + in1;                         \
-    out1 = in2 + in3;                         \
-}
-#define ADD3(in0, in1, in2, in3, in4, in5,  \
-             out0, out1, out2)              \
-{                                           \
-    ADD2(in0, in1, in2, in3, out0, out1);   \
-    out2 = in4 + in5;                       \
-}
-#define ADD4(in0, in1, in2, in3, in4, in5, in6, in7,  \
-             out0, out1, out2, out3)                  \
-{                                                     \
-    ADD2(in0, in1, in2, in3, out0, out1);             \
-    ADD2(in4, in5, in6, in7, out2, out3);             \
-}
-
-#define ILVR_B2(RTYPE, in0, in1, in2, in3, out0, out1)      \
-{                                                           \
-    out0 = (RTYPE) __msa_ilvr_b((v16i8) in0, (v16i8) in1);  \
-    out1 = (RTYPE) __msa_ilvr_b((v16i8) in2, (v16i8) in3);  \
-}
-#define ILVR_B2_SH(...) ILVR_B2(v8i16, __VA_ARGS__)
-
-#define HSUB_UB2(RTYPE, in0, in1, out0, out1)                 \
-{                                                             \
-    out0 = (RTYPE) __msa_hsub_u_h((v16u8) in0, (v16u8) in0);  \
-    out1 = (RTYPE) __msa_hsub_u_h((v16u8) in1, (v16u8) in1);  \
-}
-#define HSUB_UB2_SH(...) HSUB_UB2(v8i16, __VA_ARGS__)
-
-#define SLDI_B2_0(RTYPE, in0, in1, out0, out1, slide_val)                 \
-{                                                                         \
-    v16i8 zero_m = { 0 };                                                 \
-    out0 = (RTYPE) __msa_sldi_b((v16i8) zero_m, (v16i8) in0, slide_val);  \
-    out1 = (RTYPE) __msa_sldi_b((v16i8) zero_m, (v16i8) in1, slide_val);  \
-}
-#define SLDI_B2_0_UB(...) SLDI_B2_0(v16u8, __VA_ARGS__)
-
-#define SLDI_B3_0(RTYPE, in0, in1, in2, out0, out1, out2,  slide_val)     \
-{                                                                         \
-    v16i8 zero_m = { 0 };                                                 \
-    SLDI_B2_0(RTYPE, in0, in1, out0, out1, slide_val);                    \
-    out2 = (RTYPE) __msa_sldi_b((v16i8) zero_m, (v16i8) in2, slide_val);  \
-}
-#define SLDI_B3_0_UB(...) SLDI_B3_0(v16u8, __VA_ARGS__)
-
-#define ILVEV_W2(RTYPE, in0, in1, in2, in3, out0, out1)      \
-{                                                            \
-    out0 = (RTYPE) __msa_ilvev_w((v4i32) in1, (v4i32) in0);  \
-    out1 = (RTYPE) __msa_ilvev_w((v4i32) in3, (v4i32) in2);  \
-}
-#define ILVEV_W2_UB(...) ILVEV_W2(v16u8, __VA_ARGS__)
-
-#define ADD_ABS_H3(RTYPE, in0, in1, in2, out0, out1, out2)  \
-{                                                           \
-    RTYPE zero = {0};                                       \
-                                                            \
-    out0 = __msa_add_a_h((v8i16) zero, in0);                \
-    out1 = __msa_add_a_h((v8i16) zero, in1);                \
-    out2 = __msa_add_a_h((v8i16) zero, in2);                \
-}
-#define ADD_ABS_H3_SH(...) ADD_ABS_H3(v8i16, __VA_ARGS__)
-
-#define VSHF_B2(RTYPE, in0, in1, in2, in3, mask0, mask1, out0, out1)       \
-{                                                                          \
-    out0 = (RTYPE) __msa_vshf_b((v16i8) mask0, (v16i8) in1, (v16i8) in0);  \
-    out1 = (RTYPE) __msa_vshf_b((v16i8) mask1, (v16i8) in3, (v16i8) in2);  \
-}
-#define VSHF_B2_UB(...) VSHF_B2(v16u8, __VA_ARGS__)
-
-#define CMP_AND_SELECT(inp0, inp1, inp2, inp3, inp4, inp5, out0)              \
-{                                                                             \
-   v8i16 _sel_h0, _sel_h1;                                                    \
-   v16u8 _sel_b0, _sel_b1;                                                    \
-   _sel_h0 = (v8i16) __msa_clt_u_h((v8u16) inp1, (v8u16) inp0);               \
-   _sel_b0 = (v16u8) __msa_pckev_b((v16i8) _sel_h0, (v16i8) _sel_h0);         \
-   inp0 = (v8i16) __msa_bmnz_v((v16u8) inp0, (v16u8) inp1, (v16u8) _sel_h0);  \
-   inp4 = (v16u8) __msa_bmnz_v(inp3, inp4, _sel_b0);                          \
-   _sel_h1 = (v8i16) __msa_clt_u_h((v8u16) inp2, (v8u16) inp0);               \
-   _sel_b1 = (v16u8) __msa_pckev_b((v16i8) _sel_h1, (v16i8) _sel_h1);         \
-   inp4 = (v16u8) __msa_bmnz_v(inp4, inp5, _sel_b1);                          \
-   out0 += inp4;                                                              \
-}
-
-void png_read_filter_row_up_msa(png_row_infop row_info, png_bytep row,
-                                png_const_bytep prev_row)
-{
-   png_size_t i, cnt, cnt16, cnt32;
-   png_size_t istop = row_info->rowbytes;
-   png_bytep rp = row;
-   png_const_bytep pp = prev_row;
-   v16u8 src0, src1, src2, src3, src4, src5, src6, src7;
-
-   for (i = 0; i < (istop >> 6); i++)
-   {
-      LD_UB4(rp, 16, src0, src1, src2, src3);
-      LD_UB4(pp, 16, src4, src5, src6, src7);
-      pp += 64;
-
-	  ADD4(src0, src4, src1, src5, src2, src6, src3, src7,
-	       src0, src1, src2, src3);
-
-      ST_UB4(src0, src1, src2, src3, rp, 16);
-      rp += 64;
-   }
-
-   if (istop & 0x3F)
-   {
-      cnt32 = istop & 0x20;
-      cnt16 = istop & 0x10;
-      cnt = istop & 0xF;
-
-      if(cnt32)
-      {
-         if (cnt16 && cnt)
-         {
-            LD_UB4(rp, 16, src0, src1, src2, src3);
-            LD_UB4(pp, 16, src4, src5, src6, src7);
-
-            ADD4(src0, src4, src1, src5, src2, src6, src3, src7,
-	             src0, src1, src2, src3);
-
-            ST_UB4(src0, src1, src2, src3, rp, 16);
-            rp += 64;
-         }
-         else if (cnt16 || cnt)
-         {
-            LD_UB2(rp, 16, src0, src1);
-            LD_UB2(pp, 16, src4, src5);
-            pp += 32;
-            src2 = LD_UB(rp + 32);
-            src6 = LD_UB(pp);
-
-            ADD3(src0, src4, src1, src5, src2, src6, src0, src1, src2);
-
-            ST_UB2(src0, src1, rp, 16);
-            rp += 32;
-            ST_UB(src2, rp);
-            rp += 16;
-         }
-         else
-         {
-            LD_UB2(rp, 16, src0, src1);
-            LD_UB2(pp, 16, src4, src5);
-
-			ADD2(src0, src4, src1, src5, src0, src1);
-
-            ST_UB2(src0, src1, rp, 16);
-            rp += 32;
-         }
-      }
-      else if (cnt16 && cnt)
-      {
-         LD_UB2(rp, 16, src0, src1);
-         LD_UB2(pp, 16, src4, src5);
-
-         ADD2(src0, src4, src1, src5, src0, src1);
-
-         ST_UB2(src0, src1, rp, 16);
-         rp += 32;
-      }
-      else if (cnt16 || cnt)
-      {
-         src0 = LD_UB(rp);
-         src4 = LD_UB(pp);
-         pp += 16;
-
-         src0 += src4;
-
-         ST_UB(src0, rp);
-         rp += 16;
-      }
-   }
-}
-
-void png_read_filter_row_sub4_msa(png_row_infop row_info, png_bytep row,
-                                  png_const_bytep prev_row)
-{
-   png_size_t count;
-   png_size_t istop = row_info->rowbytes;
-   png_bytep src = row;
-   png_bytep nxt = row + 4;
-   int32_t inp0;
-   v16u8 src0, src1, src2, src3, src4;
-   v16u8 dst0, dst1;
-   v16u8 zero = { 0 };
-
-   istop -= 4;
-
-   inp0 = LW(src);
-   src += 4;
-   src0 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp0);
-
-   for (count = 0; count < istop; count += 16)
-   {
-      src1 = LD_UB(src);
-      src += 16;
-
-      src2 = (v16u8) __msa_sldi_b((v16i8) zero, (v16i8) src1, 4);
-      src3 = (v16u8) __msa_sldi_b((v16i8) zero, (v16i8) src1, 8);
-      src4 = (v16u8) __msa_sldi_b((v16i8) zero, (v16i8) src1, 12);
-      src1 += src0;
-      src2 += src1;
-      src3 += src2;
-      src4 += src3;
-      src0 = src4;
-      ILVEV_W2_UB(src1, src2, src3, src4, dst0, dst1);
-      dst0 = (v16u8) __msa_pckev_d((v2i64) dst1, (v2i64) dst0);
-
-      ST_UB(dst0, nxt);
-      nxt += 16;
-   }
-}
-
-void png_read_filter_row_sub3_msa(png_row_infop row_info, png_bytep row,
-                                  png_const_bytep prev_row)
-{
-   png_size_t count;
-   png_size_t istop = row_info->rowbytes;
-   png_bytep src = row;
-   png_bytep nxt = row + 3;
-   int64_t out0;
-   int32_t inp0, out1;
-   v16u8 src0, src1, src2, src3, src4, dst0, dst1;
-   v16u8 zero = { 0 };
-   v16i8 mask0 = { 0, 1, 2, 16, 17, 18, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
-   v16i8 mask1 = { 0, 1, 2, 3, 4, 5, 16, 17, 18, 19, 20, 21, 0, 0, 0, 0 };
-
-   istop -= 3;
-
-   inp0 = LW(src);
-   src += 3;
-   src0 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp0);
-
-   for (count = 0; count < istop; count += 12)
-   {
-      src1 = LD_UB(src);
-      src += 12;
-
-      src2 = (v16u8) __msa_sldi_b((v16i8) zero, (v16i8) src1, 3);
-      src3 = (v16u8) __msa_sldi_b((v16i8) zero, (v16i8) src1, 6);
-      src4 = (v16u8) __msa_sldi_b((v16i8) zero, (v16i8) src1, 9);
-      src1 += src0;
-      src2 += src1;
-      src3 += src2;
-      src4 += src3;
-      src0 = src4;
-      VSHF_B2_UB(src1, src2, src3, src4, mask0, mask0, dst0, dst1);
-      dst0 = (v16u8) __msa_vshf_b(mask1, (v16i8) dst1, (v16i8) dst0);
-      out0 = __msa_copy_s_d((v2i64) dst0, 0);
-      out1 = __msa_copy_s_w((v4i32) dst0, 2);
-
-      SD(out0, nxt);
-      nxt += 8;
-      SW(out1, nxt);
-      nxt += 4;
-   }
-}
-
-void png_read_filter_row_avg4_msa(png_row_infop row_info, png_bytep row,
-                                  png_const_bytep prev_row)
-{
-   png_size_t i;
-   png_bytep src = row;
-   png_bytep nxt = row;
-   png_const_bytep pp = prev_row;
-   png_size_t istop = row_info->rowbytes - 4;
-   int32_t inp0, inp1, out0;
-   v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8, src9, dst0, dst1;
-   v16u8 zero = { 0 };
-
-   inp0 = LW(pp);
-   pp += 4;
-   inp1 = LW(src);
-   src += 4;
-   src0 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp0);
-   src1 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp1);
-   src0 = (v16u8) MSA_SRLI_B(src0, 1);
-   src1 += src0;
-   out0 = __msa_copy_s_w((v4i32) src1, 0);
-   SW(out0, nxt);
-   nxt += 4;
-
-   for (i = 0; i < istop; i += 16)
-   {
-      src2 = LD_UB(pp);
-      pp += 16;
-      src6 = LD_UB(src);
-      src += 16;
-
-      SLDI_B2_0_UB(src2, src6, src3, src7, 4);
-      SLDI_B2_0_UB(src2, src6, src4, src8, 8);
-      SLDI_B2_0_UB(src2, src6, src5, src9, 12);
-      src2 = __msa_ave_u_b(src2, src1);
-      src6 += src2;
-      src3 = __msa_ave_u_b(src3, src6);
-      src7 += src3;
-      src4 = __msa_ave_u_b(src4, src7);
-      src8 += src4;
-      src5 = __msa_ave_u_b(src5, src8);
-      src9 += src5;
-      src1 = src9;
-      ILVEV_W2_UB(src6, src7, src8, src9, dst0, dst1);
-      dst0 = (v16u8) __msa_pckev_d((v2i64) dst1, (v2i64) dst0);
-
-      ST_UB(dst0, nxt);
-      nxt += 16;
-   }
-}
-
-void png_read_filter_row_avg3_msa(png_row_infop row_info, png_bytep row,
-                                  png_const_bytep prev_row)
-{
-   png_size_t i;
-   png_bytep src = row;
-   png_bytep nxt = row;
-   png_const_bytep pp = prev_row;
-   png_size_t istop = row_info->rowbytes - 3;
-   int64_t out0;
-   int32_t inp0, inp1, out1;
-   int16_t out2;
-   v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8, src9, dst0, dst1;
-   v16u8 zero = { 0 };
-   v16i8 mask0 = { 0, 1, 2, 16, 17, 18, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
-   v16i8 mask1 = { 0, 1, 2, 3, 4, 5, 16, 17, 18, 19, 20, 21, 0, 0, 0, 0 };
-
-   inp0 = LW(pp);
-   pp += 3;
-   inp1 = LW(src);
-   src += 3;
-   src0 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp0);
-   src1 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp1);
-   src0 = (v16u8) MSA_SRLI_B(src0, 1);
-   src1 += src0;
-   out2 = __msa_copy_s_h((v8i16) src1, 0);
-   SH(out2, nxt);
-   nxt += 2;
-   nxt[0] = src1[2];
-   nxt++;
-
-   for (i = 0; i < istop; i += 12)
-   {
-      src2 = LD_UB(pp);
-      pp += 12;
-      src6 = LD_UB(src);
-      src += 12;
-
-      SLDI_B2_0_UB(src2, src6, src3, src7, 3);
-      SLDI_B2_0_UB(src2, src6, src4, src8, 6);
-      SLDI_B2_0_UB(src2, src6, src5, src9, 9);
-      src2 = __msa_ave_u_b(src2, src1);
-      src6 += src2;
-      src3 = __msa_ave_u_b(src3, src6);
-      src7 += src3;
-      src4 = __msa_ave_u_b(src4, src7);
-      src8 += src4;
-      src5 = __msa_ave_u_b(src5, src8);
-      src9 += src5;
-      src1 = src9;
-      VSHF_B2_UB(src6, src7, src8, src9, mask0, mask0, dst0, dst1);
-      dst0 = (v16u8) __msa_vshf_b(mask1, (v16i8) dst1, (v16i8) dst0);
-      out0 = __msa_copy_s_d((v2i64) dst0, 0);
-      out1 = __msa_copy_s_w((v4i32) dst0, 2);
-
-      SD(out0, nxt);
-      nxt += 8;
-      SW(out1, nxt);
-      nxt += 4;
-   }
-}
-
-void png_read_filter_row_paeth4_msa(png_row_infop row_info,
-                                    png_bytep row,
-                                    png_const_bytep prev_row)
-{
-   int32_t count, rp_end;
-   png_bytep nxt;
-   png_const_bytep prev_nxt;
-   int32_t inp0, inp1, res0;
-   v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8, src9;
-   v16u8 src10, src11, src12, src13, dst0, dst1;
-   v8i16 vec0, vec1, vec2;
-   v16u8 zero = { 0 };
-
-   nxt = row;
-   prev_nxt = prev_row;
-
-   inp0 = LW(nxt);
-   inp1 = LW(prev_nxt);
-   prev_nxt += 4;
-   src0 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp0);
-   src1 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp1);
-
-   src1 += src0;
-   res0 = __msa_copy_s_w((v4i32) src1, 0);
-
-   SW(res0, nxt);
-   nxt += 4;
-
-   /* Remainder */
-   rp_end = row_info->rowbytes - 4;
-
-   for (count = 0; count < rp_end; count += 16)
-   {
-      src2 = LD_UB(prev_nxt);
-      prev_nxt += 16;
-      src6 = LD_UB(prev_row);
-      prev_row += 16;
-      src10 = LD_UB(nxt);
-
-      SLDI_B3_0_UB(src2, src6, src10, src3, src7, src11, 4);
-      SLDI_B3_0_UB(src2, src6, src10, src4, src8, src12, 8);
-      SLDI_B3_0_UB(src2, src6, src10, src5, src9, src13, 12);
-      ILVR_B2_SH(src2, src6, src1, src6, vec0, vec1);
-      HSUB_UB2_SH(vec0, vec1, vec0, vec1);
-      vec2 = vec0 + vec1;
-      ADD_ABS_H3_SH(vec0, vec1, vec2, vec0, vec1, vec2);
-      CMP_AND_SELECT(vec0, vec1, vec2, src1, src2, src6, src10);
-      ILVR_B2_SH(src3, src7, src10, src7, vec0, vec1);
-      HSUB_UB2_SH(vec0, vec1, vec0, vec1);
-      vec2 = vec0 + vec1;
-      ADD_ABS_H3_SH(vec0, vec1, vec2, vec0, vec1, vec2);
-      CMP_AND_SELECT(vec0, vec1, vec2, src10, src3, src7, src11);
-      ILVR_B2_SH(src4, src8, src11, src8, vec0, vec1);
-      HSUB_UB2_SH(vec0, vec1, vec0, vec1);
-      vec2 = vec0 + vec1;
-      ADD_ABS_H3_SH(vec0, vec1, vec2, vec0, vec1, vec2);
-      CMP_AND_SELECT(vec0, vec1, vec2, src11, src4, src8, src12);
-      ILVR_B2_SH(src5, src9, src12, src9, vec0, vec1);
-      HSUB_UB2_SH(vec0, vec1, vec0, vec1);
-      vec2 = vec0 + vec1;
-      ADD_ABS_H3_SH(vec0, vec1, vec2, vec0, vec1, vec2);
-      CMP_AND_SELECT(vec0, vec1, vec2, src12, src5, src9, src13);
-      src1 = src13;
-      ILVEV_W2_UB(src10, src11, src12, src1, dst0, dst1);
-      dst0 = (v16u8) __msa_pckev_d((v2i64) dst1, (v2i64) dst0);
-
-      ST_UB(dst0, nxt);
-      nxt += 16;
-   }
-}
-
-void png_read_filter_row_paeth3_msa(png_row_infop row_info,
-                                    png_bytep row,
-                                    png_const_bytep prev_row)
-{
-   int32_t count, rp_end;
-   png_bytep nxt;
-   png_const_bytep prev_nxt;
-   int64_t out0;
-   int32_t inp0, inp1, out1;
-   int16_t out2;
-   v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8, src9, dst0, dst1;
-   v16u8 src10, src11, src12, src13;
-   v8i16 vec0, vec1, vec2;
-   v16u8 zero = { 0 };
-   v16i8 mask0 = { 0, 1, 2, 16, 17, 18, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
-   v16i8 mask1 = { 0, 1, 2, 3, 4, 5, 16, 17, 18, 19, 20, 21, 0, 0, 0, 0 };
-
-   nxt = row;
-   prev_nxt = prev_row;
-
-   inp0 = LW(nxt);
-   inp1 = LW(prev_nxt);
-   prev_nxt += 3;
-   src0 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp0);
-   src1 = (v16u8) __msa_insert_w((v4i32) zero, 0, inp1);
-
-   src1 += src0;
-   out2 = __msa_copy_s_h((v8i16) src1, 0);
-
-   SH(out2, nxt);
-   nxt += 2;
-   nxt[0] = src1[2];
-   nxt++;
-
-   /* Remainder */
-   rp_end = row_info->rowbytes - 3;
-
-   for (count = 0; count < rp_end; count += 12)
-   {
-      src2 = LD_UB(prev_nxt);
-      prev_nxt += 12;
-      src6 = LD_UB(prev_row);
-      prev_row += 12;
-      src10 = LD_UB(nxt);
-
-      SLDI_B3_0_UB(src2, src6, src10, src3, src7, src11, 3);
-      SLDI_B3_0_UB(src2, src6, src10, src4, src8, src12, 6);
-      SLDI_B3_0_UB(src2, src6, src10, src5, src9, src13, 9);
-      ILVR_B2_SH(src2, src6, src1, src6, vec0, vec1);
-      HSUB_UB2_SH(vec0, vec1, vec0, vec1);
-      vec2 = vec0 + vec1;
-      ADD_ABS_H3_SH(vec0, vec1, vec2, vec0, vec1, vec2);
-      CMP_AND_SELECT(vec0, vec1, vec2, src1, src2, src6, src10);
-      ILVR_B2_SH(src3, src7, src10, src7, vec0, vec1);
-      HSUB_UB2_SH(vec0, vec1, vec0, vec1);
-      vec2 = vec0 + vec1;
-      ADD_ABS_H3_SH(vec0, vec1, vec2, vec0, vec1, vec2);
-      CMP_AND_SELECT(vec0, vec1, vec2, src10, src3, src7, src11);
-      ILVR_B2_SH(src4, src8, src11, src8, vec0, vec1);
-      HSUB_UB2_SH(vec0, vec1, vec0, vec1);
-      vec2 = vec0 + vec1;
-      ADD_ABS_H3_SH(vec0, vec1, vec2, vec0, vec1, vec2);
-      CMP_AND_SELECT(vec0, vec1, vec2, src11, src4, src8, src12);
-      ILVR_B2_SH(src5, src9, src12, src9, vec0, vec1);
-      HSUB_UB2_SH(vec0, vec1, vec0, vec1);
-      vec2 = vec0 + vec1;
-      ADD_ABS_H3_SH(vec0, vec1, vec2, vec0, vec1, vec2);
-      CMP_AND_SELECT(vec0, vec1, vec2, src12, src5, src9, src13);
-      src1 = src13;
-      VSHF_B2_UB(src10, src11, src12, src13, mask0, mask0, dst0, dst1);
-      dst0 = (v16u8) __msa_vshf_b(mask1, (v16i8) dst1, (v16i8) dst0);
-      out0 = __msa_copy_s_d((v2i64) dst0, 0);
-      out1 = __msa_copy_s_w((v4i32) dst0, 2);
-
-      SD(out0, nxt);
-      nxt += 8;
-      SW(out1, nxt);
-      nxt += 4;
-   }
-}
-
-#endif /* PNG_MIPS_MSA_OPT > 0 */
-#endif /* PNG_MIPS_MSA_IMPLEMENTATION == 1 (intrinsics) */
-#endif /* READ */

mips/mips_init.c

@@ -1,129 +0,0 @@
-
-/* mips_init.c - MSA optimised filter functions
- *
- * Copyright (c) 2016 Glenn Randers-Pehrson
- * Written by Mandar Sahastrabuddhe, 2016.
- * Last changed in libpng 1.6.25 [September 1, 2016]
- *
- * This code is released under the libpng license.
- * For conditions of distribution and use, see the disclaimer
- * and license in png.h
- */
-/* Below, after checking __linux__, various non-C90 POSIX 1003.1 functions are
- * called.
- */
-#define _POSIX_SOURCE 1
-
-#include <stdio.h>
-#include "../pngpriv.h"
-
-#ifdef PNG_READ_SUPPORTED
-
-#if PNG_MIPS_MSA_OPT > 0
-#ifdef PNG_MIPS_MSA_CHECK_SUPPORTED /* Do run-time checks */
-/* WARNING: it is strongly recommended that you do not build libpng with
- * run-time checks for CPU features if at all possible.  In the case of the MIPS
- * MSA instructions there is no processor-specific way of detecting the
- * presence of the required support, therefore run-time detection is extremely
- * OS specific.
- *
- * You may set the macro PNG_MIPS_MSA_FILE to the file name of file containing
- * a fragment of C source code which defines the png_have_msa function.  There
- * are a number of implementations in contrib/mips-msa, but the only one that
- * has partial support is contrib/mips-msa/linux.c - a generic Linux
- * implementation which reads /proc/cpufino.
- */
-#ifndef PNG_MIPS_MSA_FILE
-#  ifdef __linux__
-#     define PNG_MIPS_MSA_FILE "contrib/mips-msa/linux.c"
-#  endif
-#endif
-
-#ifdef PNG_MIPS_MSA_FILE
-
-#include <signal.h> /* for sig_atomic_t */
-static int png_have_msa(png_structp png_ptr);
-#include PNG_MIPS_MSA_FILE
-
-#else  /* PNG_MIPS_MSA_FILE */
-#  error "PNG_MIPS_MSA_FILE undefined: no support for run-time MIPS MSA checks"
-#endif /* PNG_MIPS_MSA_FILE */
-#endif /* PNG_MIPS_MSA_CHECK_SUPPORTED */
-
-#ifndef PNG_ALIGNED_MEMORY_SUPPORTED
-#  error "ALIGNED_MEMORY is required; set: -DPNG_ALIGNED_MEMORY_SUPPORTED"
-#endif
-
-void
-png_init_filter_functions_msa(png_structp pp, unsigned int bpp)
-{
-   /* The switch statement is compiled in for MIPS_MSA_API, the call to
-    * png_have_msa is compiled in for MIPS_MSA_CHECK. If both are defined
-    * the check is only performed if the API has not set the MSA option on
-    * or off explicitly. In this case the check controls what happens.
-    */
-
-#ifdef PNG_MIPS_MSA_API_SUPPORTED
-   switch ((pp->options >> PNG_MIPS_MSA) & 3)
-   {
-      case PNG_OPTION_UNSET:
-         /* Allow the run-time check to execute if it has been enabled -
-          * thus both API and CHECK can be turned on.  If it isn't supported
-          * this case will fall through to the 'default' below, which just
-          * returns.
-          */
-#endif /* PNG_MIPS_MSA_API_SUPPORTED */
-#ifdef PNG_MIPS_MSA_CHECK_SUPPORTED
-         {
-            static volatile sig_atomic_t no_msa = -1; /* not checked */
-
-            if (no_msa < 0)
-               no_msa = !png_have_msa(pp);
-
-            if (no_msa)
-               return;
-         }
-#ifdef PNG_MIPS_MSA_API_SUPPORTED
-         break;
-#endif
-#endif /* PNG_MIPS_MSA_CHECK_SUPPORTED */
-
-#ifdef PNG_MIPS_MSA_API_SUPPORTED
-      default: /* OFF or INVALID */
-         return;
-
-      case PNG_OPTION_ON:
-         /* Option turned on */
-         break;
-   }
-#endif
-
-   /* IMPORTANT: any new external functions used here must be declared using
-    * PNG_INTERNAL_FUNCTION in ../pngpriv.h.  This is required so that the
-    * 'prefix' option to configure works:
-    *
-    *    ./configure --with-libpng-prefix=foobar_
-    *
-    * Verify you have got this right by running the above command, doing a build
-    * and examining pngprefix.h; it must contain a #define for every external
-    * function you add.  (Notice that this happens automatically for the
-    * initialization function.)
-    */
-   pp->read_filter[PNG_FILTER_VALUE_UP-1] = png_read_filter_row_up_msa;
-
-   if (bpp == 3)
-   {
-      pp->read_filter[PNG_FILTER_VALUE_SUB-1] = png_read_filter_row_sub3_msa;
-      pp->read_filter[PNG_FILTER_VALUE_AVG-1] = png_read_filter_row_avg3_msa;
-      pp->read_filter[PNG_FILTER_VALUE_PAETH-1] = png_read_filter_row_paeth3_msa;
-   }
-
-   else if (bpp == 4)
-   {
-      pp->read_filter[PNG_FILTER_VALUE_SUB-1] = png_read_filter_row_sub4_msa;
-      pp->read_filter[PNG_FILTER_VALUE_AVG-1] = png_read_filter_row_avg4_msa;
-      pp->read_filter[PNG_FILTER_VALUE_PAETH-1] = png_read_filter_row_paeth4_msa;
-   }
-}
-#endif /* PNG_MIPS_MSA_OPT > 0 */
-#endif /* READ */

png.5

@@ -1,4 +1,4 @@
-.TH PNG 5 "April 22, 2017"
+.TH PNG 5 "November 12, 2015"
 .SH NAME
 png \- Portable Network Graphics (PNG) format
 .SH DESCRIPTION
@@ -23,11 +23,12 @@ platforms.
 PNG specification (second edition), November 2003:
 .IP
 .br
-  <https://www.w3.org/TR/2003/REC-PNG-20031110/
+  <http://www.w3.org/TR/2003/REC-PNG-20031110/
+.LP
 PNG 1.2 specification, July 1999:
 .IP
 .br
-https://png-mng.sourceforge.io/pub/png/spec/1.2/
+http://png-mng.sourceforge.net/pub/png/spec/1.2/
 .LP
 PNG 1.0 specification, October 1996:
 .IP
@@ -35,11 +36,11 @@ PNG 1.0 specification, October 1996:
 RFC 2083
 .IP
 .br
-https://www.ietf.org/rfc/rfc2083.txt
+ftp://ds.internic.net/rfc/rfc2083.txt
 .br
 or (as a W3C Recommendation) at
 .br
-https://www.w3.org/TR/REC-png-961001
+http://www.w3.org/TR/REC-png-961001
 .SH AUTHORS
 This man page: Glenn Randers-Pehrson
 .LP
@@ -58,7 +59,7 @@ Thomas Boutell and others (png-list).
 
 .SH COPYRIGHT NOTICE
 .LP
-This man page is Copyright (c) 1998-2006 Glenn Randers-Pehrson.  See png.h
+This man page is Copyright (c) 1998-2012 Glenn Randers-Pehrson.  See png.h
 for conditions of use and distribution.
 .LP
 The PNG Specification (Second Edition) is

pngchunk.h

@@ -0,0 +1,47 @@
+
+/* pngchunk.h - PNG chunk descriptions
+ *
+ * Last changed in libpng 1.7.0 [(PENDING RELEASE)]
+ * Copyright (c) 2015 Glenn Randers-Pehrson
+ *
+ * This code is released under the libpng license.
+ * For conditions of distribution and use, see the disclaimer
+ * and license in png.h
+ *
+ * If this list is changed in any way scripts/chunkhash.c must be rebuilt and
+ * run to regenerate the lookup functions for the tables described from this
+ * list.
+ *
+ * IDAT MUST be first in the list; it must have index '0'.  The order of the
+ * remaining chunks comes from section 5.6 "Chunk ordering" in the ISO spec
+ * plus the ordering rules in the PNG extensions documnet.
+ *
+ * Keep PNG_CHUNK_BEGIN and PNG_CHUNK_END at the beginning and end.
+ */
+PNG_CHUNK_BEGIN(IDAT,  73,  68,  65,  84, within_IDAT,  after_start)
+PNG_CHUNK(      IHDR,  73,  72,  68,  82, before_start, at_start)
+PNG_CHUNK(      cHRM,  99,  72,  82,  77, before_PLTE,  after_start)
+PNG_CHUNK(      gAMA, 103,  65,  77,  65, before_PLTE,  after_start)
+PNG_CHUNK(      iCCP, 105,  67,  67,  80, before_PLTE,  after_start)
+PNG_CHUNK(      sBIT, 115,  66,  73,  84, before_PLTE,  after_start)
+PNG_CHUNK(      sRGB, 115,  82,  71,  66, before_PLTE,  after_start)
+PNG_CHUNK(      PLTE,  80,  76,  84,  69, before_PLTE,  after_start)
+PNG_CHUNK(      bKGD,  98,  75,  71,  68, before_IDAT,  after_PLTE)
+PNG_CHUNK(      hIST, 104,  73,  83,  84, before_IDAT,  after_PLTE)
+PNG_CHUNK(      tRNS, 116,  82,  78,  83, before_IDAT,  after_PLTE)
+PNG_CHUNK(      oFFs, 111,  70,  70, 115, before_IDAT,  after_start)
+PNG_CHUNK(      pCAL, 112,  67,  65,  76, before_IDAT,  after_start)
+PNG_CHUNK(      sCAL, 115,  67,  65,  76, before_IDAT,  after_start)
+PNG_CHUNK(      sTER, 115,  84,  69,  82, before_IDAT,  after_start)
+PNG_CHUNK(      pHYs, 112,  72,  89, 115, before_IDAT,  after_start)
+PNG_CHUNK(      sPLT, 115,  80,  76,  84, before_IDAT,  after_start)
+PNG_CHUNK(      tIME, 116,  73,  77,  69, before_end,   after_start)
+PNG_CHUNK(      iTXt, 105,  84,  88, 116, before_end,   after_start)
+PNG_CHUNK(      tEXt, 116,  69,  88, 116, before_end,   after_start)
+PNG_CHUNK(      zTXt, 122,  84,  88, 116, before_end,   after_start)
+PNG_CHUNK(      fRAc, 102,  82,  65,  99, before_end,   after_start)
+PNG_CHUNK(      gIFg, 103,  73,  70, 103, before_end,   after_start)
+PNG_CHUNK(      gIFt, 103,  73,  70, 116, before_end,   after_start)
+PNG_CHUNK(      gIFx, 103,  73,  70, 120, before_end,   after_start)
+PNG_CHUNK(      dSIG, 100,  83,  73,  71, before_end,   after_start)
+PNG_CHUNK_END(  IEND,  73,  69,  78,  68, before_end,   after_IDAT)

pngconf.h

@@ -1,9 +1,9 @@
 
 /* pngconf.h - machine configurable file for libpng
  *
- * libpng version 1.6.30beta02, April 22, 2017
+ * libpng version 1.7.0beta68, November 12, 2015
  *
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Copyright (c) 1998-2015 Glenn Randers-Pehrson
  * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
  * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
  *
@@ -63,7 +63,7 @@
  */
 #define PNG_CONST const /* backward compatibility only */
 
-/* This controls optimization of the reading of 16-bit and 32-bit values
+/* This controls optimization of the reading of 16 and 32 bit values
  * from PNG files.  It can be set on a per-app-file basis - it
  * just changes whether a macro is used when the function is called.
  * The library builder sets the default; if read functions are not
@@ -188,27 +188,27 @@
    * compatible with GCC or Visual C because of different calling conventions.
    */
 #  if PNG_API_RULE == 2
-   /* If this line results in an error, either because __watcall is not
-    * understood or because of a redefine just below you cannot use *this*
-    * build of the library with the compiler you are using.  *This* build was
-    * build using Watcom and applications must also be built using Watcom!
-    */
+    /* If this line results in an error, either because __watcall is not
+     * understood or because of a redefine just below you cannot use *this*
+     * build of the library with the compiler you are using.  *This* build was
+     * build using Watcom and applications must also be built using Watcom!
+     */
 #    define PNGCAPI __watcall
 #  endif
 
 #  if defined(__GNUC__) || (defined(_MSC_VER) && (_MSC_VER >= 800))
 #    define PNGCAPI __cdecl
 #    if PNG_API_RULE == 1
-   /* If this line results in an error __stdcall is not understood and
-    * PNG_API_RULE should not have been set to '1'.
-    */
+       /* If this line results in an error __stdcall is not understood and
+        * PNG_API_RULE should not have been set to '1'.
+        */
 #      define PNGAPI __stdcall
 #    endif
 #  else
-   /* An older compiler, or one not detected (erroneously) above,
-    * if necessary override on the command line to get the correct
-    * variants for the compiler.
-    */
+    /* An older compiler, or one not detected (erroneously) above,
+     * if necessary override on the command line to get the correct
+     * variants for the compiler.
+     */
 #    ifndef PNGCAPI
 #      define PNGCAPI _cdecl
 #    endif
@@ -225,10 +225,10 @@
 
 #  if (defined(_MSC_VER) && _MSC_VER < 800) ||\
       (defined(__BORLANDC__) && __BORLANDC__ < 0x500)
-   /* older Borland and MSC
-    * compilers used '__export' and required this to be after
-    * the type.
-    */
+    /* older Borland and MSC
+     * compilers used '__export' and required this to be after
+     * the type.
+     */
 #    ifndef PNG_EXPORT_TYPE
 #      define PNG_EXPORT_TYPE(type) type PNG_IMPEXP
 #    endif
@@ -244,9 +244,9 @@
 #  if (defined(__IBMC__) || defined(__IBMCPP__)) && defined(__OS2__)
 #    define PNGAPI _System
 #  else /* !Windows/x86 && !OS/2 */
-   /* Use the defaults, or define PNG*API on the command line (but
-    * this will have to be done for every compile!)
-    */
+    /* Use the defaults, or define PNG*API on the command line (but
+     * this will have to be done for every compile!)
+     */
 #  endif /* other system, !OS/2 */
 #endif /* !Windows/x86 */
 
@@ -267,7 +267,7 @@
  */
 #ifndef PNG_IMPEXP
 #  if defined(PNG_USE_DLL) && defined(PNG_DLL_IMPORT)
-   /* This forces use of a DLL, disallowing static linking */
+     /* This forces use of a DLL, disallowing static linking */
 #    define PNG_IMPEXP PNG_DLL_IMPORT
 #  endif
 
@@ -295,11 +295,11 @@
     * table entries, so we discard it here.  See the .dfn files in the
     * scripts directory.
     */
-
 #ifndef PNG_EXPORTA
-#  define PNG_EXPORTA(ordinal, type, name, args, attributes) \
-      PNG_FUNCTION(PNG_EXPORT_TYPE(type), (PNGAPI name), PNGARG(args), \
-      PNG_LINKAGE_API attributes)
+
+#  define PNG_EXPORTA(ordinal, type, name, args, attributes)\
+      PNG_FUNCTION(PNG_EXPORT_TYPE(type),(PNGAPI name),PNGARG(args), \
+        extern attributes)
 #endif
 
 /* ANSI-C (C90) does not permit a macro to be invoked with an empty argument,
@@ -307,7 +307,7 @@
  */
 #define PNG_EMPTY /*empty list*/
 
-#define PNG_EXPORT(ordinal, type, name, args) \
+#define PNG_EXPORT(ordinal, type, name, args)\
    PNG_EXPORTA(ordinal, type, name, args, PNG_EMPTY)
 
 /* Use PNG_REMOVED to comment out a removed interface. */
@@ -340,7 +340,7 @@
    * less efficient code.
    */
 #  if defined(__clang__) && defined(__has_attribute)
-   /* Clang defines both __clang__ and __GNUC__. Check __clang__ first. */
+     /* Clang defines both __clang__ and __GNUC__. Check __clang__ first. */
 #    if !defined(PNG_USE_RESULT) && __has_attribute(__warn_unused_result__)
 #      define PNG_USE_RESULT __attribute__((__warn_unused_result__))
 #    endif
@@ -445,6 +445,14 @@
 #  define PNG_RESTRICT    /* The C99 "restrict" feature */
 #endif
 
+#ifndef PNG_FP_EXPORT     /* A floating point API. */
+#  ifdef PNG_FLOATING_POINT_SUPPORTED
+#     define PNG_FP_EXPORT(ordinal, type, name, args)\
+         PNG_EXPORT(ordinal, type, name, args);
+#  else                   /* No floating point APIs */
+#     define PNG_FP_EXPORT(ordinal, type, name, args)
+#  endif
+#endif
 #ifndef PNG_FP_EXPORT     /* A floating point API. */
 #  ifdef PNG_FLOATING_POINT_SUPPORTED
 #     define PNG_FP_EXPORT(ordinal, type, name, args)\
@@ -480,7 +488,7 @@
 #if CHAR_BIT == 8 && UCHAR_MAX == 255
    typedef unsigned char png_byte;
 #else
-#  error "libpng requires 8-bit bytes"
+#  error "libpng requires 8 bit bytes"
 #endif
 
 #if INT_MIN == -32768 && INT_MAX == 32767
@@ -488,7 +496,7 @@
 #elif SHRT_MIN == -32768 && SHRT_MAX == 32767
    typedef short png_int_16;
 #else
-#  error "libpng requires a signed 16-bit type"
+#  error "libpng requires a signed 16 bit type"
 #endif
 
 #if UINT_MAX == 65535
@@ -496,7 +504,7 @@
 #elif USHRT_MAX == 65535
    typedef unsigned short png_uint_16;
 #else
-#  error "libpng requires an unsigned 16-bit type"
+#  error "libpng requires an unsigned 16 bit type"
 #endif
 
 #if INT_MIN < -2147483646 && INT_MAX > 2147483646
@@ -504,19 +512,21 @@
 #elif LONG_MIN < -2147483646 && LONG_MAX > 2147483646
    typedef long int png_int_32;
 #else
-#  error "libpng requires a signed 32-bit (or more) type"
+#  error "libpng requires a signed 32 bit (or more) type"
 #endif
 
-#if UINT_MAX > 4294967294U
+#if UINT_MAX > 4294967294
    typedef unsigned int png_uint_32;
-#elif ULONG_MAX > 4294967294U
+#elif ULONG_MAX > 4294967294
    typedef unsigned long int png_uint_32;
 #else
-#  error "libpng requires an unsigned 32-bit (or more) type"
+#  error "libpng requires an unsigned 32 bit (or more) type"
 #endif
 
 /* Prior to 1.6.0 it was possible to disable the use of size_t, 1.6.0, however,
  * requires an ISOC90 compiler and relies on consistent behavior of sizeof.
+ *
+ * DEPRECATED: don't use these types, instead use size_t and ptrdiff_t.
  */
 typedef size_t png_size_t;
 typedef ptrdiff_t png_ptrdiff_t;
@@ -537,13 +547,13 @@ typedef ptrdiff_t png_ptrdiff_t;
 #  endif
 #endif
 
-/* png_alloc_size_t is guaranteed to be no smaller than png_size_t, and no
- * smaller than png_uint_32.  Casts from png_size_t or png_uint_32 to
+/* png_alloc_size_t is guaranteed to be no smaller than size_t, and no
+ * smaller than png_uint_32.  Casts from size_t or png_uint_32 to
  * png_alloc_size_t are not necessary; in fact, it is recommended not to use
  * them at all so that the compiler can complain when something turns out to be
  * problematic.
  *
- * Casts in the other direction (from png_alloc_size_t to png_size_t or
+ * Casts in the other direction (from png_alloc_size_t to size_t or
  * png_uint_32) should be explicitly applied; however, we do not expect to
  * encounter practical situations that require such conversions.
  *
@@ -553,7 +563,7 @@ typedef ptrdiff_t png_ptrdiff_t;
 #ifdef PNG_SMALL_SIZE_T
    typedef png_uint_32 png_alloc_size_t;
 #else
-   typedef png_size_t png_alloc_size_t;
+   typedef size_t png_alloc_size_t;
 #endif
 
 /* Prior to 1.6.0 libpng offered limited support for Microsoft C compiler
@@ -589,8 +599,8 @@ typedef char                  * png_charp;
 typedef const char            * png_const_charp;
 typedef png_fixed_point       * png_fixed_point_p;
 typedef const png_fixed_point * png_const_fixed_point_p;
-typedef png_size_t            * png_size_tp;
-typedef const png_size_t      * png_const_size_tp;
+typedef size_t                * png_size_tp;
+typedef const size_t          * png_const_size_tp;
 
 #ifdef PNG_STDIO_SUPPORTED
 typedef FILE            * png_FILE_p;

pngdebug.h

@@ -2,7 +2,7 @@
 /* pngdebug.h - Debugging macros for libpng, also used in pngtest.c
  *
  * Last changed in libpng 1.6.8 [December 19, 2013]
- * Copyright (c) 1998-2002,2004,2006-2013 Glenn Randers-Pehrson
+ * Copyright (c) 1998-2013 Glenn Randers-Pehrson
  * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
  * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
  *
@@ -36,13 +36,13 @@
 /* These settings control the formatting of messages in png.c and pngerror.c */
 /* Moved to pngdebug.h at 1.5.0 */
 #  ifndef PNG_LITERAL_SHARP
-#    define PNG_LITERAL_SHARP 0x23
+#    define PNG_LITERAL_SHARP 0x23U
 #  endif
 #  ifndef PNG_LITERAL_LEFT_SQUARE_BRACKET
-#    define PNG_LITERAL_LEFT_SQUARE_BRACKET 0x5b
+#    define PNG_LITERAL_LEFT_SQUARE_BRACKET 0x5bU
 #  endif
 #  ifndef PNG_LITERAL_RIGHT_SQUARE_BRACKET
-#    define PNG_LITERAL_RIGHT_SQUARE_BRACKET 0x5d
+#    define PNG_LITERAL_RIGHT_SQUARE_BRACKET 0x5dU
 #  endif
 #  ifndef PNG_STRING_NEWLINE
 #    define PNG_STRING_NEWLINE "\n"

pngerror.c

@@ -1,8 +1,8 @@
 
 /* pngerror.c - stub functions for i/o and memory allocation
  *
- * Last changed in libpng 1.6.26 [October 20, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.7.0 [(PENDING RELEASE)]
+ * Copyright (c) 1998-2015 Glenn Randers-Pehrson
  * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
  * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
  *
@@ -17,6 +17,7 @@
  */
 
 #include "pngpriv.h"
+#define PNG_SRC_FILE PNG_SRC_FILE_pngerror
 
 #if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED)
 
@@ -26,7 +27,7 @@ static PNG_FUNCTION(void, png_default_error,PNGARG((png_const_structrp png_ptr,
 #ifdef PNG_WARNINGS_SUPPORTED
 static void /* PRIVATE */
 png_default_warning PNGARG((png_const_structrp png_ptr,
-    png_const_charp warning_message));
+   png_const_charp warning_message));
 #endif /* WARNINGS */
 
 /* This function is called whenever there is a fatal error.  This function
@@ -37,7 +38,7 @@ png_default_warning PNGARG((png_const_structrp png_ptr,
 #ifdef PNG_ERROR_TEXT_SUPPORTED
 PNG_FUNCTION(void,PNGAPI
 png_error,(png_const_structrp png_ptr, png_const_charp error_message),
-    PNG_NORETURN)
+   PNG_NORETURN)
 {
 #ifdef PNG_ERROR_NUMBERS_SUPPORTED
    char msg[16];
@@ -65,18 +66,18 @@ png_error,(png_const_structrp png_ptr, png_const_charp error_message),
 
             else
                error_message += offset;
-         }
-
-         else
-         {
-            if ((png_ptr->flags & PNG_FLAG_STRIP_ERROR_TEXT) != 0)
-            {
-               msg[0] = '0';
-               msg[1] = '\0';
-               error_message = msg;
-            }
-         }
       }
+
+      else
+      {
+         if ((png_ptr->flags & PNG_FLAG_STRIP_ERROR_TEXT) != 0)
+         {
+            msg[0] = '0';
+            msg[1] = '\0';
+            error_message = msg;
+         }
+       }
+     }
    }
 #endif
    if (png_ptr != NULL && png_ptr->error_fn != NULL)
@@ -110,7 +111,7 @@ png_err,(png_const_structrp png_ptr),PNG_NORETURN)
  */
 size_t
 png_safecat(png_charp buffer, size_t bufsize, size_t pos,
-    png_const_charp string)
+   png_const_charp string)
 {
    if (buffer != NULL && pos < bufsize)
    {
@@ -129,9 +130,10 @@ png_safecat(png_charp buffer, size_t bufsize, size_t pos,
  * and end pointer (which should point just *beyond* the end of the buffer!)
  * Returns the pointer to the start of the formatted string.
  */
+#define PNG_HAVE_FORMAT_NUMBER /* for the code below */
 png_charp
 png_format_number(png_const_charp start, png_charp end, int format,
-    png_alloc_size_t number)
+   png_alloc_size_t number)
 {
    int count = 0;    /* number of digits output */
    int mincount = 1; /* minimum number required */
@@ -233,7 +235,7 @@ png_warning(png_const_structrp png_ptr, png_const_charp warning_message)
    }
    if (png_ptr != NULL && png_ptr->warning_fn != NULL)
       (*(png_ptr->warning_fn))(png_constcast(png_structrp,png_ptr),
-          warning_message + offset);
+         warning_message + offset);
    else
       png_default_warning(png_ptr, warning_message + offset);
 }
@@ -245,7 +247,7 @@ png_warning(png_const_structrp png_ptr, png_const_charp warning_message)
  */
 void
 png_warning_parameter(png_warning_parameters p, int number,
-    png_const_charp string)
+   png_const_charp string)
 {
    if (number > 0 && number <= PNG_WARNING_PARAMETER_COUNT)
       (void)png_safecat(p[number-1], (sizeof p[number-1]), 0, string);
@@ -253,7 +255,7 @@ png_warning_parameter(png_warning_parameters p, int number,
 
 void
 png_warning_parameter_unsigned(png_warning_parameters p, int number, int format,
-    png_alloc_size_t value)
+   png_alloc_size_t value)
 {
    char buffer[PNG_NUMBER_BUFFER_SIZE];
    png_warning_parameter(p, number, PNG_FORMAT_NUMBER(buffer, format, value));
@@ -261,7 +263,7 @@ png_warning_parameter_unsigned(png_warning_parameters p, int number, int format,
 
 void
 png_warning_parameter_signed(png_warning_parameters p, int number, int format,
-    png_int_32 value)
+   png_int_32 value)
 {
    png_alloc_size_t u;
    png_charp str;
@@ -282,7 +284,7 @@ png_warning_parameter_signed(png_warning_parameters p, int number, int format,
 
 void
 png_formatted_warning(png_const_structrp png_ptr, png_warning_parameters p,
-    png_const_charp message)
+   png_const_charp message)
 {
    /* The internal buffer is just 192 bytes - enough for all our messages,
     * overflow doesn't happen because this code checks!  If someone figures
@@ -364,8 +366,7 @@ png_benign_error(png_const_structrp png_ptr, png_const_charp error_message)
    if ((png_ptr->flags & PNG_FLAG_BENIGN_ERRORS_WARN) != 0)
    {
 #     ifdef PNG_READ_SUPPORTED
-         if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0 &&
-            png_ptr->chunk_name != 0)
+         if (png_ptr->read_struct && png_ptr->chunk_name != 0)
             png_chunk_warning(png_ptr, error_message);
          else
 #     endif
@@ -375,8 +376,7 @@ png_benign_error(png_const_structrp png_ptr, png_const_charp error_message)
    else
    {
 #     ifdef PNG_READ_SUPPORTED
-         if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0 &&
-            png_ptr->chunk_name != 0)
+         if (png_ptr->read_struct && png_ptr->chunk_name != 0)
             png_chunk_error(png_ptr, error_message);
          else
 #     endif
@@ -391,10 +391,10 @@ png_benign_error(png_const_structrp png_ptr, png_const_charp error_message)
 void /* PRIVATE */
 png_app_warning(png_const_structrp png_ptr, png_const_charp error_message)
 {
-   if ((png_ptr->flags & PNG_FLAG_APP_WARNINGS_WARN) != 0)
-      png_warning(png_ptr, error_message);
-   else
-      png_error(png_ptr, error_message);
+  if ((png_ptr->flags & PNG_FLAG_APP_WARNINGS_WARN) != 0)
+     png_warning(png_ptr, error_message);
+  else
+     png_error(png_ptr, error_message);
 
 #  ifndef PNG_ERROR_TEXT_SUPPORTED
       PNG_UNUSED(error_message)
@@ -404,10 +404,10 @@ png_app_warning(png_const_structrp png_ptr, png_const_charp error_message)
 void /* PRIVATE */
 png_app_error(png_const_structrp png_ptr, png_const_charp error_message)
 {
-   if ((png_ptr->flags & PNG_FLAG_APP_ERRORS_WARN) != 0)
-      png_warning(png_ptr, error_message);
-   else
-      png_error(png_ptr, error_message);
+  if ((png_ptr->flags & PNG_FLAG_APP_ERRORS_WARN) != 0)
+     png_warning(png_ptr, error_message);
+  else
+     png_error(png_ptr, error_message);
 
 #  ifndef PNG_ERROR_TEXT_SUPPORTED
       PNG_UNUSED(error_message)
@@ -419,10 +419,14 @@ png_app_error(png_const_structrp png_ptr, png_const_charp error_message)
 #if defined(PNG_WARNINGS_SUPPORTED) || \
    (defined(PNG_READ_SUPPORTED) && defined(PNG_ERROR_TEXT_SUPPORTED))
 /* These utilities are used internally to build an error message that relates
- * to the current chunk.  The chunk name comes from png_ptr->chunk_name,
- * which is used to prefix the message.  The message is limited in length
- * to 63 bytes. The name characters are output as hex digits wrapped in []
- * if the character is invalid.
+ * to the current chunk.  The chunk name comes from png_ptr->chunk_name unless
+ * png_ptr->zowner is set in which case that is used in preference.  This is
+ * used to prefix the message.  The message is limited in length to 63 bytes.
+ * The name characters are output as hex digits wrapped in [] if the character
+ * is invalid.
+ *
+ * Using 'zowner' means that IDAT errors at the end of the IDAT stream are still
+ * reported as from the IDAT chunks.
  */
 #define isnonalpha(c) ((c) < 65 || (c) > 122 || ((c) > 90 && (c) < 97))
 static PNG_CONST char png_digit[16] = {
@@ -434,9 +438,12 @@ static void /* PRIVATE */
 png_format_buffer(png_const_structrp png_ptr, png_charp buffer, png_const_charp
     error_message)
 {
-   png_uint_32 chunk_name = png_ptr->chunk_name;
+   png_uint_32 chunk_name = png_ptr->zowner;
    int iout = 0, ishift = 24;
 
+   if (chunk_name == 0)
+      chunk_name = png_ptr->chunk_name;
+
    while (ishift >= 0)
    {
       int c = (int)(chunk_name >> ishift) & 0xff;
@@ -452,7 +459,7 @@ png_format_buffer(png_const_structrp png_ptr, png_charp buffer, png_const_charp
 
       else
       {
-         buffer[iout++] = (char)c;
+         buffer[iout++] = png_check_char(png_ptr, c);
       }
    }
 
@@ -478,7 +485,7 @@ png_format_buffer(png_const_structrp png_ptr, png_charp buffer, png_const_charp
 #if defined(PNG_READ_SUPPORTED) && defined(PNG_ERROR_TEXT_SUPPORTED)
 PNG_FUNCTION(void,PNGAPI
 png_chunk_error,(png_const_structrp png_ptr, png_const_charp error_message),
-    PNG_NORETURN)
+   PNG_NORETURN)
 {
    char msg[18+PNG_MAX_ERROR_TEXT];
    if (png_ptr == NULL)
@@ -528,17 +535,14 @@ png_chunk_benign_error(png_const_structrp png_ptr, png_const_charp
 #endif /* READ */
 
 void /* PRIVATE */
-png_chunk_report(png_const_structrp png_ptr, png_const_charp message, int error)
+(png_chunk_report)(png_const_structrp png_ptr, png_const_charp message,
+   int error)
 {
-#  ifndef PNG_WARNINGS_SUPPORTED
-      PNG_UNUSED(message)
-#  endif
-
    /* This is always supported, but for just read or just write it
     * unconditionally does the right thing.
     */
 #  if defined(PNG_READ_SUPPORTED) && defined(PNG_WRITE_SUPPORTED)
-      if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0)
+      if (png_ptr->read_struct)
 #  endif
 
 #  ifdef PNG_READ_SUPPORTED
@@ -546,13 +550,16 @@ png_chunk_report(png_const_structrp png_ptr, png_const_charp message, int error)
          if (error < PNG_CHUNK_ERROR)
             png_chunk_warning(png_ptr, message);
 
-         else
+         else if (error < PNG_CHUNK_FATAL)
             png_chunk_benign_error(png_ptr, message);
+
+         else
+            png_chunk_error(png_ptr, message);
       }
 #  endif
 
 #  if defined(PNG_READ_SUPPORTED) && defined(PNG_WRITE_SUPPORTED)
-      else if ((png_ptr->mode & PNG_IS_READ_STRUCT) == 0)
+      else if (!png_ptr->read_struct)
 #  endif
 
 #  ifdef PNG_WRITE_SUPPORTED
@@ -560,20 +567,33 @@ png_chunk_report(png_const_structrp png_ptr, png_const_charp message, int error)
          if (error < PNG_CHUNK_WRITE_ERROR)
             png_app_warning(png_ptr, message);
 
-         else
+         else if (error < PNG_CHUNK_FATAL)
             png_app_error(png_ptr, message);
+
+         else
+            png_error(png_ptr, message);
       }
 #  endif
+
+#  ifndef PNG_ERROR_TEXT_SUPPORTED
+      PNG_UNUSED(message)
+#  endif
 }
 
 #ifdef PNG_ERROR_TEXT_SUPPORTED
-#ifdef PNG_FLOATING_POINT_SUPPORTED
+
+#if defined(PNG_FLOATING_POINT_SUPPORTED) && \
+   (defined(PNG_gAMA_SUPPORTED) || defined(PNG_cHRM_SUPPORTED) || \
+   defined(PNG_sCAL_SUPPORTED) || defined(PNG_READ_BACKGROUND_SUPPORTED) || \
+   defined(PNG_READ_RGB_TO_GRAY_SUPPORTED)) || \
+   (defined(PNG_FLOATING_ARITHMETIC_SUPPORTED) &&\
+   defined(PNG_sCAL_SUPPORTED))
 PNG_FUNCTION(void,
 png_fixed_error,(png_const_structrp png_ptr, png_const_charp name),PNG_NORETURN)
 {
 #  define fixed_message "fixed point overflow in "
 #  define fixed_message_ln ((sizeof fixed_message)-1)
-   unsigned int  iin;
+   int  iin;
    char msg[fixed_message_ln+PNG_MAX_ERROR_TEXT];
    memcpy(msg, fixed_message, fixed_message_ln);
    iin = 0;
@@ -620,7 +640,7 @@ png_set_longjmp_fn(png_structrp png_ptr, png_longjmp_ptr longjmp_fn,
       else
       {
          png_ptr->jmp_buf_ptr = png_voidcast(jmp_buf *,
-             png_malloc_warn(png_ptr, jmp_buf_size));
+            png_malloc_warn(png_ptr, jmp_buf_size));
 
          if (png_ptr->jmp_buf_ptr == NULL)
             return NULL; /* new NULL return on OOM */
@@ -641,9 +661,9 @@ png_set_longjmp_fn(png_structrp png_ptr, png_longjmp_ptr longjmp_fn,
             /* This is an internal error in libpng: somehow we have been left
              * with a stack allocated jmp_buf when the application regained
              * control.  It's always possible to fix this up, but for the moment
-             * this is a png_error because that makes it easy to detect.
+             * this is an affirm because that makes it easy to detect.
              */
-            png_error(png_ptr, "Libpng jmp_buf still allocated");
+            impossible("Libpng jmp_buf still allocated");
             /* png_ptr->jmp_buf_ptr = &png_ptr->jmp_buf_local; */
          }
       }
@@ -709,7 +729,7 @@ png_free_jmpbuf(png_structrp png_ptr)
  */
 static PNG_FUNCTION(void /* PRIVATE */,
 png_default_error,(png_const_structrp png_ptr, png_const_charp error_message),
-    PNG_NORETURN)
+   PNG_NORETURN)
 {
 #ifdef PNG_CONSOLE_IO_SUPPORTED
 #ifdef PNG_ERROR_NUMBERS_SUPPORTED
@@ -768,11 +788,15 @@ png_longjmp,(png_const_structrp png_ptr, int val),PNG_NORETURN)
 
    /* If control reaches this point, png_longjmp() must not return. The only
     * choice is to terminate the whole process (or maybe the thread); to do
-    * this the ANSI-C abort() function is used unless a different method is
+    * this the ANSI-C abort() function is used unless a different method is 
     * implemented by overriding the default configuration setting for
-    * PNG_ABORT().
+    * PNG_ABORT (see scripts/pnglibconf.dfa).
+    *
+    * API change: prior to 1.7.0 PNG_ABORT was invoked as a function type macro
+    * with no arguments 'PNG_ABORT();', in 1.7.0 this is changed to a simple
+    * macro that is defined in the configuration.
     */
-   PNG_ABORT();
+   PNG_ABORT
 }
 
 #ifdef PNG_WARNINGS_SUPPORTED
@@ -869,7 +893,7 @@ png_set_strip_error_numbers(png_structrp png_ptr, png_uint_32 strip_mode)
    if (png_ptr != NULL)
    {
       png_ptr->flags &=
-         ((~(PNG_FLAG_STRIP_ERROR_NUMBERS |
+         ((PNG_BIC_MASK(PNG_FLAG_STRIP_ERROR_NUMBERS |
          PNG_FLAG_STRIP_ERROR_TEXT))&strip_mode);
    }
 }
@@ -883,7 +907,7 @@ png_set_strip_error_numbers(png_structrp png_ptr, png_uint_32 strip_mode)
     */
 PNG_FUNCTION(void /* PRIVATE */, (PNGCBAPI
 png_safe_error),(png_structp png_nonconst_ptr, png_const_charp error_message),
-    PNG_NORETURN)
+   PNG_NORETURN)
 {
    const png_const_structrp png_ptr = png_nonconst_ptr;
    png_imagep image = png_voidcast(png_imagep, png_ptr->error_ptr);
@@ -906,7 +930,7 @@ png_safe_error),(png_structp png_nonconst_ptr, png_const_charp error_message),
       /* Missing longjmp buffer, the following is to help debugging: */
       {
          size_t pos = png_safecat(image->message, (sizeof image->message), 0,
-             "bad longjmp: ");
+            "bad longjmp: ");
          png_safecat(image->message, (sizeof image->message), pos,
              error_message);
       }
@@ -960,4 +984,258 @@ png_safe_execute(png_imagep image_in, int (*function)(png_voidp), png_voidp arg)
    return result;
 }
 #endif /* SIMPLIFIED READ || SIMPLIFIED_WRITE */
+
+/* Affirms: minimal code in 'STABLE' builds to return control to the
+ * application via png_error(), more verbose code followed by PNG_ABORT for
+ * all other builds to ensure that internal errors are detected.
+ *
+ * The code always produces a message if it is possible, regardless of the
+ * setting of PNG_ERROR_TEXT_SUPPORTED, except that in stable builds
+ * PNG_ERROR_TEXT_SUPPORTED is honored.  See pngpriv.h for the calculation of
+ * the two control macros PNG_RELEASE_BUILD (don't abort; stable build or rc)
+ * and PNG_AFFIRM_TEXT (output text.)
+ */
+#if PNG_AFFIRM_TEXT
+#  ifdef PNG_HAVE_FORMAT_NUMBER
+static size_t
+png_affirm_number(png_charp buffer, size_t bufsize, size_t pos,
+   unsigned int number, int format)
+{
+   char numbuf[PNG_NUMBER_BUFFER_SIZE];
+   return png_safecat(buffer, bufsize, pos,
+      png_format_number(numbuf, numbuf + sizeof numbuf, format, number));
+}
+#  define affirm_number(a,b,c,d,e) png_affirm_number(a,b,c,d,e)
+#  else /* !HAVE_FORMAT_NUMBER */
+static size_t
+png_affirm_number(png_charp buffer, size_t bufsize, size_t pos,
+   unsigned int number)
+{
+   /* binhex it; highly non-portable, assumes the ASCII character set, but
+    * if warnings are turned off then it is unlikely the text will get read
+    * anyway.  This binhex variant is (48+val), where 'val' is the next 6
+    * bits of the number, so it starts as '0' (for 0) and ends at 'I' for
+    * 63.  The number is wrapped in {}, so 0 comes out as '{}' and 9 comes
+    * out as '{9}' and so on.
+    */
+   char numbuf[32];
+   int i = sizeof numbuf;
+
+   numbuf[--i] = 0;
+   numbuf[--i] = '}';
+
+   do
+   {
+      if (number > 0)
+         numbuf[--i] = (char)/*SAFE*/((number & 63) + 48), number >>= 6;
+      else
+      {
+         numbuf[--i] = '{';
+         break;
+      }
+   }
+   while (i > 0);
+
+   return png_safecat(buffer, bufsize, pos, numbuf+i);
+}
+#  define affirm_number(a,b,c,d,e) png_affirm_number(a,b,c,d)
+#endif /* !HAVE_FORMAT_NUMBER */
+
+static void
+affirm_text(png_charp buffer, size_t bufsize,
+   param_deb(png_const_charp condition) unsigned int position)
+{
+  /* Format the 'position' number and output:
+   *
+   *  "<file> <line>: affirm 'condition' failed\n"
+   *  " libpng version <version> - <date>\n"
+   *  " translated __DATE__ __TIME__"
+   *
+   * In the STABLE versions the output is the same for the last two lines
+   * but the first line becomes:
+   *
+   *  "<position>: affirm failed"
+   *
+   * If there is no number formatting the numbers just get replaced by
+   * some binhex (see the utility above).
+   */
+  size_t pos = 0;
+
+# if PNG_RELEASE_BUILD /* no 'condition' parameter: minimal text */
+     pos = affirm_number(buffer, bufsize, pos, position, PNG_NUMBER_FORMAT_x);
+     pos = png_safecat(buffer, bufsize, pos, ": affirm failed");
+# else /* !STABLE */
+     /* Break down 'position' into a file name and a line number: */
+     {
+#        define PNG_apply(f) { #f "\0", PNG_SRC_FILE_ ## f },
+#        define PNG_end      { "", PNG_SRC_FILE_LAST }
+         static struct {
+             char         filename[28]; /* GCC checks this size */
+             unsigned int base;
+         } fileinfo[] = { PNG_FILES };
+#        undef PNG_apply
+#        undef PNG_end
+
+         unsigned int i;
+         png_const_charp filename;
+
+         /* Do 'nfiles' this way to avoid problems with g++ where it whines
+          * about (size_t) being larger than (int), even though this is a
+          * compile time constant:
+          */
+#        define nfiles ((sizeof fileinfo)/(sizeof (fileinfo[0])))
+         for (i=0; i < nfiles && position > fileinfo[i].base; ++i) {}
+
+         if (i == 0 || i > nfiles)
+             filename = "UNKNOWN";
+         else
+         {
+             filename = fileinfo[i-1].filename;
+             position -= fileinfo[i-1].base;
+         }
+#        undef nfiles
+
+         pos = png_safecat(buffer, bufsize, pos, filename);
+         pos = png_safecat(buffer, bufsize, pos, ".c ");
+         pos = affirm_number(buffer, bufsize, pos, position,
+            PNG_NUMBER_FORMAT_u);
+     }
+
+     pos = png_safecat(buffer, bufsize, pos, ": affirm '");
+     pos = png_safecat(buffer, bufsize, pos, condition);
+     pos = png_safecat(buffer, bufsize, pos, "' failed\n");
+# endif /* !STABLE */
+
+  pos = png_safecat(buffer, bufsize, pos, PNG_HEADER_VERSION_STRING);
+  pos = png_safecat(buffer, bufsize, pos,
+     " translated " __DATE__ " " __TIME__);
+}
+
+#define affirm_text(b, c, p)\
+   do {\
+      (affirm_text)(b, sizeof b, param_deb(c) (p));\
+   } while (0)
+
+#endif /* AFFIRM_TEXT */
+
+PNG_FUNCTION(void,png_affirm,(png_const_structrp png_ptr,
+   param_deb(png_const_charp condition) unsigned int position),PNG_NORETURN)
+{
+#  if PNG_AFFIRM_TEXT
+      char   buffer[512];
+
+      affirm_text(buffer, condition, position);
+#  else /* !AFFIRM_TEXT */
+      PNG_UNUSED(position)
+#     if !PNG_RELEASE_BUILD
+         PNG_UNUSED(condition)
+#     endif
+#  endif /* AFFIRM_TEXT */
+
+   /* Now in STABLE do a png_error, but in other builds output the message
+    * (if possible) then abort (PNG_ABORT).
+    */
+#  if PNG_RELEASE_BUILD
+      png_error(png_ptr, buffer/*macro parameter used only if ERROR_TEXT*/);
+#  else /* !AFFIRM_ERROR */
+      /* Use console IO if possible; this is because there is no guarantee that
+       * the app 'warning' will output anything.  For certain the simplified
+       * API implementation just copies the message (truncated) to the image
+       * message buffer, which makes debugging much more difficult.
+       *
+       * Note that it is possible that neither WARNINGS nor CONSOLE_IO are
+       * supported; in that case no text will be output (and PNG_AFFIRM_TEXT
+       * will be false.)
+       */
+#     ifdef PNG_CONSOLE_IO_SUPPORTED
+         fprintf(stderr, "%s\n", buffer);
+         PNG_UNUSED(png_ptr)
+#     elif defined PNG_WARNINGS_SUPPORTED
+         if (png_ptr != NULL && png_ptr->warning_fn != NULL)
+            png_ptr->warning_fn(png_constcast(png_structrp, png_ptr), buffer);
+         /* else no way to output the text */
+#     else
+         PNG_UNUSED(png_ptr)
+#     endif
+
+      PNG_ABORT
+#  endif /* AFFIRM_ERROR */
+}
+
+#if !PNG_RELEASE_BUILD
+void /* PRIVATE */
+png_handled_affirm(png_const_structrp png_ptr, png_const_charp message,
+   unsigned int position)
+{
+#  if PNG_RELEASE_BUILD
+      /* testing in RC: we want to return control to the caller, so do not
+       * use png_affirm.
+       */
+      char   buffer[512];
+
+      affirm_text(buffer, message, position);
+
+#     ifdef PNG_CONSOLE_IO_SUPPORTED
+         fprintf(stderr, "%s\n", buffer);
+#     elif defined PNG_WARNINGS_SUPPORTED
+         if (png_ptr != NULL && png_ptr->warning_fn != NULL)
+            png_ptr->warning_fn(png_constcast(png_structrp, png_ptr), buffer);
+         /* else no way to output the text */
+#     else
+         PNG_UNUSED(png_ptr)
+#     endif
+#  else
+      png_affirm(png_ptr, message, position);
+#  endif
+}
+#endif /* !RELEASE_BUILD */
+
+#ifdef PNG_RANGE_CHECK_SUPPORTED
+/* The character/byte checking APIs. These do their own calls to png_affirm
+ * because the caller provides the position.
+ */
+unsigned int /* PRIVATE */
+png_bit_affirm(png_const_structrp png_ptr, unsigned int position,
+   unsigned int u, unsigned int bits)
+{
+   /* The following avoids overflow errors even if 'bits' is 16 or 32: */
+   if (u <= (1U << bits)-1U)
+       return u;
+
+   png_affirm(png_ptr, param_deb("(bit field) range") position);
+}
+
+char /* PRIVATE */
+png_char_affirm(png_const_structrp png_ptr, unsigned int position, int c)
+{
+   if (c >= CHAR_MIN && c <= CHAR_MAX)
+       return (char)/*SAFE*/c;
+
+   png_affirm(png_ptr, param_deb("(char) range") position);
+}
+
+png_byte /* PRIVATE */
+png_byte_affirm(png_const_structrp png_ptr, unsigned int position, int b)
+{
+   /* For the type png_byte the limits.h values are ignored and we check
+    * against the values PNG expects to store in a byte:
+    */
+   if (b >= 0 && b <= 255)
+       return (png_byte)/*SAFE*/b;
+
+   png_affirm(png_ptr, param_deb("PNG byte range") position);
+}
+
+#if INT_MAX >= 65535
+png_uint_16 /* PRIVATE */
+png_u16_affirm(png_const_structrp png_ptr, unsigned int position, int b)
+{
+   /* Check against the PNG 16-bit limit, as with png_byte. */
+   if (b >= 0 && b <= 65535)
+       return (png_uint_16)/*SAFE*/b;
+
+   png_affirm(png_ptr, param_deb("PNG 16-bit range") position);
+}
+#endif /* INT_MAX >= 65535 */
+#endif /* RANGE_CHECK */
 #endif /* READ || WRITE */

pngget.c

@@ -1,8 +1,8 @@
 
 /* pngget.c - retrieval of values from info struct
  *
- * Last changed in libpng 1.6.26 [October 20, 2016]
- * Copyright (c) 1998-2002,2004,2006-2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.7.0 [(PENDING RELEASE)]
+ * Copyright (c) 1998-2015 Glenn Randers-Pehrson
  * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
  * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
  *
@@ -13,6 +13,7 @@
  */
 
 #include "pngpriv.h"
+#define PNG_SRC_FILE PNG_SRC_FILE_pngget
 
 #if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED)
 
@@ -26,16 +27,18 @@ png_get_valid(png_const_structrp png_ptr, png_const_inforp info_ptr,
    return(0);
 }
 
-png_size_t PNGAPI
+png_alloc_size_t PNGAPI
 png_get_rowbytes(png_const_structrp png_ptr, png_const_inforp info_ptr)
 {
    if (png_ptr != NULL && info_ptr != NULL)
-      return(info_ptr->rowbytes);
+      return png_calc_rowbytes(png_ptr,
+         PNG_FORMAT_CHANNELS(info_ptr->format) * info_ptr->bit_depth,
+         info_ptr->width);
 
-   return(0);
+   return 0;
 }
 
-#ifdef PNG_INFO_IMAGE_SUPPORTED
+#ifdef PNG_READ_PNG_SUPPORTED
 png_bytepp PNGAPI
 png_get_rows(png_const_structrp png_ptr, png_const_inforp info_ptr)
 {
@@ -79,7 +82,8 @@ png_byte PNGAPI
 png_get_color_type(png_const_structrp png_ptr, png_const_inforp info_ptr)
 {
    if (png_ptr != NULL && info_ptr != NULL)
-      return info_ptr->color_type;
+      return png_check_byte(png_ptr,
+            PNG_COLOR_TYPE_FROM_FORMAT(info_ptr->format));
 
    return (0);
 }
@@ -316,7 +320,7 @@ png_get_y_offset_pixels(png_const_structrp png_ptr, png_const_inforp info_ptr)
 static png_uint_32
 ppi_from_ppm(png_uint_32 ppm)
 {
-#if 0
+#if 0 /*NOT USED*/
    /* The conversion is *(2.54/100), in binary (32 digits):
     * .00000110100000001001110101001001
     */
@@ -338,7 +342,7 @@ ppi_from_ppm(png_uint_32 ppm)
    png_fixed_point result;
    if (ppm <= PNG_UINT_31_MAX && png_muldiv(&result, (png_int_32)ppm, 127,
        5000) != 0)
-      return (png_uint_32)result;
+      return result;
 
    /* Overflow. */
    return 0;
@@ -365,14 +369,25 @@ png_get_y_pixels_per_inch(png_const_structrp png_ptr, png_const_inforp info_ptr)
 
 #ifdef PNG_FIXED_POINT_SUPPORTED
 static png_fixed_point
-png_fixed_inches_from_microns(png_const_structrp png_ptr, png_int_32 microns)
+png_fixed_inches_from_microns(png_const_structrp png_ptr,
+   png_int_32 microns)
 {
    /* Convert from metres * 1,000,000 to inches * 100,000, meters to
     * inches is simply *(100/2.54), so we want *(10/2.54) == 500/127.
     * Notice that this can overflow - a warning is output and 0 is
     * returned.
     */
-   return png_muldiv_warn(png_ptr, microns, 500, 127);
+   png_fixed_point result;
+
+   if (png_muldiv(&result, microns, 500, 127) != 0)
+      return result;
+
+   /* TODO: review this, png_error might be better. */
+   png_warning(png_ptr, "inch to microns overflow");
+   return 0;
+#  ifndef PNG_WARNINGS_SUPPORTED
+      PNG_UNUSED(png_ptr)
+#  endif
 }
 
 png_fixed_point PNGAPI
@@ -382,9 +397,7 @@ png_get_x_offset_inches_fixed(png_const_structrp png_ptr,
    return png_fixed_inches_from_microns(png_ptr,
        png_get_x_offset_microns(png_ptr, info_ptr));
 }
-#endif
 
-#ifdef PNG_FIXED_POINT_SUPPORTED
 png_fixed_point PNGAPI
 png_get_y_offset_inches_fixed(png_const_structrp png_ptr,
     png_const_inforp info_ptr)
@@ -392,7 +405,7 @@ png_get_y_offset_inches_fixed(png_const_structrp png_ptr,
    return png_fixed_inches_from_microns(png_ptr,
        png_get_y_offset_microns(png_ptr, info_ptr));
 }
-#endif
+#endif /* FIXED_POINT */
 
 #ifdef PNG_FLOATING_POINT_SUPPORTED
 float PNGAPI
@@ -456,18 +469,18 @@ png_get_pHYs_dpi(png_const_structrp png_ptr, png_const_inforp info_ptr,
    return (retval);
 }
 #endif /* pHYs */
-#endif /* INCH_CONVERSIONS */
+#endif  /* INCH_CONVERSIONS */
 
 /* png_get_channels really belongs in here, too, but it's been around longer */
 
-#endif /* EASY_ACCESS */
+#endif  /* EASY_ACCESS */
 
 
 png_byte PNGAPI
 png_get_channels(png_const_structrp png_ptr, png_const_inforp info_ptr)
 {
    if (png_ptr != NULL && info_ptr != NULL)
-      return(info_ptr->channels);
+      return png_check_byte(png_ptr, PNG_FORMAT_CHANNELS(info_ptr->format));
 
    return (0);
 }
@@ -486,7 +499,7 @@ png_get_signature(png_const_structrp png_ptr, png_const_inforp info_ptr)
 #ifdef PNG_bKGD_SUPPORTED
 png_uint_32 PNGAPI
 png_get_bKGD(png_const_structrp png_ptr, png_inforp info_ptr,
-    png_color_16p *background)
+   png_color_16p *background)
 {
    if (png_ptr != NULL && info_ptr != NULL &&
        (info_ptr->valid & PNG_INFO_bKGD) != 0 &&
@@ -526,28 +539,28 @@ png_get_cHRM(png_const_structrp png_ptr, png_const_inforp info_ptr,
 
       if (white_x != NULL)
          *white_x = png_float(png_ptr,
-             info_ptr->colorspace.end_points_xy.whitex, "cHRM white X");
+            info_ptr->colorspace.end_points_xy.whitex, "cHRM white X");
       if (white_y != NULL)
          *white_y = png_float(png_ptr,
-             info_ptr->colorspace.end_points_xy.whitey, "cHRM white Y");
+            info_ptr->colorspace.end_points_xy.whitey, "cHRM white Y");
       if (red_x != NULL)
          *red_x = png_float(png_ptr, info_ptr->colorspace.end_points_xy.redx,
-             "cHRM red X");
+            "cHRM red X");
       if (red_y != NULL)
          *red_y = png_float(png_ptr, info_ptr->colorspace.end_points_xy.redy,
-             "cHRM red Y");
+            "cHRM red Y");
       if (green_x != NULL)
          *green_x = png_float(png_ptr,
-             info_ptr->colorspace.end_points_xy.greenx, "cHRM green X");
+            info_ptr->colorspace.end_points_xy.greenx, "cHRM green X");
       if (green_y != NULL)
          *green_y = png_float(png_ptr,
-             info_ptr->colorspace.end_points_xy.greeny, "cHRM green Y");
+            info_ptr->colorspace.end_points_xy.greeny, "cHRM green Y");
       if (blue_x != NULL)
          *blue_x = png_float(png_ptr, info_ptr->colorspace.end_points_xy.bluex,
-             "cHRM blue X");
+            "cHRM blue X");
       if (blue_y != NULL)
          *blue_y = png_float(png_ptr, info_ptr->colorspace.end_points_xy.bluey,
-             "cHRM blue Y");
+            "cHRM blue Y");
       return (PNG_INFO_cHRM);
    }
 
@@ -556,42 +569,42 @@ png_get_cHRM(png_const_structrp png_ptr, png_const_inforp info_ptr,
 
 png_uint_32 PNGAPI
 png_get_cHRM_XYZ(png_const_structrp png_ptr, png_const_inforp info_ptr,
-    double *red_X, double *red_Y, double *red_Z, double *green_X,
-    double *green_Y, double *green_Z, double *blue_X, double *blue_Y,
-    double *blue_Z)
+   double *red_X, double *red_Y, double *red_Z, double *green_X,
+   double *green_Y, double *green_Z, double *blue_X, double *blue_Y,
+   double *blue_Z)
 {
    if (png_ptr != NULL && info_ptr != NULL &&
-       (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0)
+      (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0)
    {
       png_debug1(1, "in %s retrieval function", "cHRM_XYZ(float)");
 
       if (red_X != NULL)
          *red_X = png_float(png_ptr, info_ptr->colorspace.end_points_XYZ.red_X,
-             "cHRM red X");
+            "cHRM red X");
       if (red_Y != NULL)
          *red_Y = png_float(png_ptr, info_ptr->colorspace.end_points_XYZ.red_Y,
-             "cHRM red Y");
+            "cHRM red Y");
       if (red_Z != NULL)
          *red_Z = png_float(png_ptr, info_ptr->colorspace.end_points_XYZ.red_Z,
-             "cHRM red Z");
+            "cHRM red Z");
       if (green_X != NULL)
          *green_X = png_float(png_ptr,
-             info_ptr->colorspace.end_points_XYZ.green_X, "cHRM green X");
+            info_ptr->colorspace.end_points_XYZ.green_X, "cHRM green X");
       if (green_Y != NULL)
          *green_Y = png_float(png_ptr,
-             info_ptr->colorspace.end_points_XYZ.green_Y, "cHRM green Y");
+            info_ptr->colorspace.end_points_XYZ.green_Y, "cHRM green Y");
       if (green_Z != NULL)
          *green_Z = png_float(png_ptr,
-             info_ptr->colorspace.end_points_XYZ.green_Z, "cHRM green Z");
+            info_ptr->colorspace.end_points_XYZ.green_Z, "cHRM green Z");
       if (blue_X != NULL)
          *blue_X = png_float(png_ptr,
-             info_ptr->colorspace.end_points_XYZ.blue_X, "cHRM blue X");
+            info_ptr->colorspace.end_points_XYZ.blue_X, "cHRM blue X");
       if (blue_Y != NULL)
          *blue_Y = png_float(png_ptr,
-             info_ptr->colorspace.end_points_XYZ.blue_Y, "cHRM blue Y");
+            info_ptr->colorspace.end_points_XYZ.blue_Y, "cHRM blue Y");
       if (blue_Z != NULL)
          *blue_Z = png_float(png_ptr,
-             info_ptr->colorspace.end_points_XYZ.blue_Z, "cHRM blue Z");
+            info_ptr->colorspace.end_points_XYZ.blue_Z, "cHRM blue Z");
       return (PNG_INFO_cHRM);
    }
 
@@ -681,8 +694,8 @@ png_get_gAMA_fixed(png_const_structrp png_ptr, png_const_inforp info_ptr,
    png_debug1(1, "in %s retrieval function", "gAMA");
 
    if (png_ptr != NULL && info_ptr != NULL &&
-       (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_GAMMA) != 0 &&
-       file_gamma != NULL)
+      (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_GAMMA) != 0 &&
+      file_gamma != NULL)
    {
       *file_gamma = info_ptr->colorspace.gamma;
       return (PNG_INFO_gAMA);
@@ -704,7 +717,7 @@ png_get_gAMA(png_const_structrp png_ptr, png_const_inforp info_ptr,
       file_gamma != NULL)
    {
       *file_gamma = png_float(png_ptr, info_ptr->colorspace.gamma,
-          "png_get_gAMA");
+         "png_get_gAMA");
       return (PNG_INFO_gAMA);
    }
 
@@ -812,7 +825,7 @@ png_get_IHDR(png_const_structrp png_ptr, png_const_inforp info_ptr,
        *bit_depth = info_ptr->bit_depth;
 
    if (color_type != NULL)
-       *color_type = info_ptr->color_type;
+       *color_type = PNG_COLOR_TYPE_FROM_FORMAT(info_ptr->format);
 
    if (compression_type != NULL)
       *compression_type = info_ptr->compression_type;
@@ -823,16 +836,7 @@ png_get_IHDR(png_const_structrp png_ptr, png_const_inforp info_ptr,
    if (interlace_type != NULL)
       *interlace_type = info_ptr->interlace_type;
 
-   /* This is redundant if we can be sure that the info_ptr values were all
-    * assigned in png_set_IHDR().  We do the check anyhow in case an
-    * application has ignored our advice not to mess with the members
-    * of info_ptr directly.
-    */
-   png_check_IHDR(png_ptr, info_ptr->width, info_ptr->height,
-       info_ptr->bit_depth, info_ptr->color_type, info_ptr->interlace_type,
-       info_ptr->compression_type, info_ptr->filter_type);
-
-   return (1);
+   return 1;
 }
 
 #ifdef PNG_oFFs_SUPPORTED
@@ -884,9 +888,8 @@ png_get_pCAL(png_const_structrp png_ptr, png_inforp info_ptr,
 #endif
 
 #ifdef PNG_sCAL_SUPPORTED
-#  ifdef PNG_FIXED_POINT_SUPPORTED
-#    if defined(PNG_FLOATING_ARITHMETIC_SUPPORTED) || \
-         defined(PNG_FLOATING_POINT_SUPPORTED)
+#  ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED
+#     ifdef PNG_FIXED_POINT_SUPPORTED
 png_uint_32 PNGAPI
 png_get_sCAL_fixed(png_const_structrp png_ptr, png_const_inforp info_ptr,
     int *unit, png_fixed_point *width, png_fixed_point *height)
@@ -901,15 +904,15 @@ png_get_sCAL_fixed(png_const_structrp png_ptr, png_const_inforp info_ptr,
        */
       *width = png_fixed(png_ptr, atof(info_ptr->scal_s_width), "sCAL width");
       *height = png_fixed(png_ptr, atof(info_ptr->scal_s_height),
-          "sCAL height");
+         "sCAL height");
       return (PNG_INFO_sCAL);
    }
 
    return(0);
 }
-#    endif /* FLOATING_ARITHMETIC */
-#  endif /* FIXED_POINT */
-#  ifdef PNG_FLOATING_POINT_SUPPORTED
+#     endif /* FIXED_POINT */
+
+#     ifdef PNG_FLOATING_POINT_SUPPORTED
 png_uint_32 PNGAPI
 png_get_sCAL(png_const_structrp png_ptr, png_const_inforp info_ptr,
     int *unit, double *width, double *height)
@@ -925,7 +928,9 @@ png_get_sCAL(png_const_structrp png_ptr, png_const_inforp info_ptr,
 
    return(0);
 }
-#  endif /* FLOATING POINT */
+#     endif /* FLOATING POINT */
+#  endif /* FLOATING_ARITHMETIC */
+
 png_uint_32 PNGAPI
 png_get_sCAL_s(png_const_structrp png_ptr, png_const_inforp info_ptr,
     int *unit, png_charpp width, png_charpp height)
@@ -1069,7 +1074,7 @@ png_get_tRNS(png_const_structrp png_ptr, png_inforp info_ptr,
    {
       png_debug1(1, "in %s retrieval function", "tRNS");
 
-      if (info_ptr->color_type == PNG_COLOR_TYPE_PALETTE)
+      if ((info_ptr->format & PNG_FORMAT_FLAG_COLORMAP) != 0)
       {
          if (trans_alpha != NULL)
          {
@@ -1081,7 +1086,7 @@ png_get_tRNS(png_const_structrp png_ptr, png_inforp info_ptr,
             *trans_color = &(info_ptr->trans_color);
       }
 
-      else /* if (info_ptr->color_type != PNG_COLOR_TYPE_PALETTE) */
+      else /* if (info_ptr->format not colormapped */
       {
          if (trans_color != NULL)
          {
@@ -1115,15 +1120,17 @@ png_get_unknown_chunks(png_const_structrp png_ptr, png_inforp info_ptr,
       return info_ptr->unknown_chunks_num;
    }
 
-   return (0);
+   return 0;
 }
 #endif
 
 #ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED
 png_byte PNGAPI
-png_get_rgb_to_gray_status (png_const_structrp png_ptr)
+png_get_rgb_to_gray_status(png_const_structrp png_ptr)
 {
-   return (png_byte)(png_ptr ? png_ptr->rgb_to_gray_status : 0);
+   if (png_ptr)
+      return png_ptr->rgb_to_gray_status;
+   return 0;
 }
 #endif
 
@@ -1131,7 +1138,9 @@ png_get_rgb_to_gray_status (png_const_structrp png_ptr)
 png_voidp PNGAPI
 png_get_user_chunk_ptr(png_const_structrp png_ptr)
 {
-   return (png_ptr ? png_ptr->user_chunk_ptr : NULL);
+   if (png_ptr)
+      return png_ptr->user_chunk_ptr;
+   return NULL;
 }
 #endif
 
@@ -1142,19 +1151,19 @@ png_get_compression_buffer_size(png_const_structrp png_ptr)
       return 0;
 
 #ifdef PNG_WRITE_SUPPORTED
-   if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0)
+      if (png_ptr->read_struct)
 #endif
    {
 #ifdef PNG_SEQUENTIAL_READ_SUPPORTED
-      return png_ptr->IDAT_read_size;
+         return png_ptr->IDAT_read_size;
 #else
-      return PNG_IDAT_READ_SIZE;
+         return PNG_IDAT_READ_SIZE;
 #endif
    }
 
 #ifdef PNG_WRITE_SUPPORTED
-   else
-      return png_ptr->zbuffer_size;
+      else
+         return png_ptr->zbuffer_size;
 #endif
 }
 
@@ -1202,18 +1211,4 @@ png_get_io_chunk_type (png_const_structrp png_ptr)
    return png_ptr->chunk_name;
 }
 #endif /* IO_STATE */
-
-#ifdef PNG_CHECK_FOR_INVALID_INDEX_SUPPORTED
-#  ifdef PNG_GET_PALETTE_MAX_SUPPORTED
-int PNGAPI
-png_get_palette_max(png_const_structp png_ptr, png_const_infop info_ptr)
-{
-   if (png_ptr != NULL && info_ptr != NULL)
-      return png_ptr->num_palette_max;
-
-   return (-1);
-}
-#  endif
-#endif
-
 #endif /* READ || WRITE */

pnginfo.h

@@ -2,7 +2,7 @@
 /* pnginfo.h - header file for PNG reference library
  *
  * Last changed in libpng 1.6.1 [March 28, 2013]
- * Copyright (c) 1998-2002,2004,2006-2013 Glenn Randers-Pehrson
+ * Copyright (c) 1998-2013 Glenn Randers-Pehrson
  * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
  * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
  *
@@ -11,7 +11,7 @@
  * and license in png.h
  */
 
- /* png_info is a structure that holds the information in a PNG file so
+/* png_info is a structure that holds the information in a PNG file so
  * that the application can find out the characteristics of the image.
  * If you are reading the file, this structure will tell you what is
  * in the PNG file.  If you are writing the file, fill in the information
@@ -57,25 +57,17 @@ struct png_info_def
    /* The following are necessary for every PNG file */
    png_uint_32 width;  /* width of image in pixels (from IHDR) */
    png_uint_32 height; /* height of image in pixels (from IHDR) */
-   png_uint_32 valid;  /* valid chunk data (see PNG_INFO_ below) */
-   png_size_t rowbytes; /* bytes needed to hold an untransformed row */
+   unsigned int valid;  /* valid chunk data (see PNG_INFO_ in png.h) */
    png_colorp palette;      /* array of color values (valid & PNG_INFO_PLTE) */
-   png_uint_16 num_palette; /* number of color entries in "palette" (PLTE) */
-   png_uint_16 num_trans;   /* number of transparent palette color (tRNS) */
-   png_byte bit_depth;      /* 1, 2, 4, 8, or 16 bits/channel (from IHDR) */
-   png_byte color_type;     /* see PNG_COLOR_TYPE_ below (from IHDR) */
+   unsigned int num_palette:9; /* number of color entries in "palette" (PLTE) */
+   unsigned int num_trans:9;   /* number of transparent palette color (tRNS) */
+   unsigned int bit_depth:6;   /* 1, 2, 4, 8, 16 or 32 bits/channel */
+   unsigned int format:PNG_RF_BITS; /* row format; see png_struct.h */
    /* The following three should have been named *_method not *_type */
    png_byte compression_type; /* must be PNG_COMPRESSION_TYPE_BASE (IHDR) */
    png_byte filter_type;    /* must be PNG_FILTER_TYPE_BASE (from IHDR) */
    png_byte interlace_type; /* One of PNG_INTERLACE_NONE, PNG_INTERLACE_ADAM7 */
 
-   /* The following are set by png_set_IHDR, called from the application on
-    * write, but the are never actually used by the write code.
-    */
-   png_byte channels;       /* number of data channels per pixel (1, 2, 3, 4) */
-   png_byte pixel_depth;    /* number of bits per pixel */
-   png_byte spare_byte;     /* to align the data, and for future use */
-
 #ifdef PNG_READ_SUPPORTED
    /* This is never set during write */
    png_byte signature[8];   /* magic bytes read by libpng from start of file */
@@ -223,7 +215,7 @@ defined(PNG_READ_BACKGROUND_SUPPORTED)
    /* Storage for unknown chunks that the library doesn't recognize. */
    png_unknown_chunkp unknown_chunks;
 
-   /* The type of this field is limited by the type of
+   /* The type of this field is limited by the type of 
     * png_struct::user_chunk_cache_max, else overflow can occur.
     */
    int                unknown_chunks_num;

pngmem.c

@@ -1,8 +1,8 @@
 
 /* pngmem.c - stub functions for memory allocation
  *
- * Last changed in libpng 1.6.26 [October 20, 2016]
- * Copyright (c) 1998-2002,2004,2006-2014,2016 Glenn Randers-Pehrson
+ * Last changed in libpng 1.7.0 [(PENDING RELEASE)]
+ * Copyright (c) 1998-2015 Glenn Randers-Pehrson
  * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger)
  * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.)
  *
@@ -18,6 +18,7 @@
  */
 
 #include "pngpriv.h"
+#define PNG_SRC_FILE PNG_SRC_FILE_pngmem
 
 #if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED)
 /* Free a png_struct */
@@ -66,7 +67,7 @@ png_calloc,(png_const_structrp png_ptr, png_alloc_size_t size),PNG_ALLOCATED)
  */
 PNG_FUNCTION(png_voidp /* PRIVATE */,
 png_malloc_base,(png_const_structrp png_ptr, png_alloc_size_t size),
-    PNG_ALLOCATED)
+   PNG_ALLOCATED)
 {
    /* Moved to png_malloc_base from png_malloc_default in 1.6.0; the DOS
     * allocators have also been removed in 1.6.0, so any 16-bit system now has
@@ -107,9 +108,9 @@ png_malloc_base,(png_const_structrp png_ptr, png_alloc_size_t size),
  */
 static png_voidp
 png_malloc_array_checked(png_const_structrp png_ptr, int nelements,
-    size_t element_size)
+   size_t element_size)
 {
-   png_alloc_size_t req = (png_alloc_size_t)nelements; /* known to be > 0 */
+   png_alloc_size_t req = nelements; /* known to be > 0 */
 
    if (req <= PNG_SIZE_MAX/element_size)
       return png_malloc_base(png_ptr, req * element_size);
@@ -120,22 +121,19 @@ png_malloc_array_checked(png_const_structrp png_ptr, int nelements,
 
 PNG_FUNCTION(png_voidp /* PRIVATE */,
 png_malloc_array,(png_const_structrp png_ptr, int nelements,
-    size_t element_size),PNG_ALLOCATED)
+   size_t element_size),PNG_ALLOCATED)
 {
-   if (nelements <= 0 || element_size == 0)
-      png_error(png_ptr, "internal error: array alloc");
-
+   affirm(nelements > 0 && element_size > 0);
    return png_malloc_array_checked(png_ptr, nelements, element_size);
 }
 
 PNG_FUNCTION(png_voidp /* PRIVATE */,
-png_realloc_array,(png_const_structrp png_ptr, png_const_voidp old_array,
-    int old_elements, int add_elements, size_t element_size),PNG_ALLOCATED)
+png_realloc_array,(png_structrp png_ptr, png_const_voidp old_array,
+   int old_elements, int add_elements, size_t element_size),PNG_ALLOCATED)
 {
    /* These are internal errors: */
-   if (add_elements <= 0 || element_size == 0 || old_elements < 0 ||
-      (old_array == NULL && old_elements > 0))
-      png_error(png_ptr, "internal error: array realloc");
+   affirm(add_elements > 0 && element_size > 0 && old_elements >= 0 &&
+      (old_array != NULL || old_elements == 0));
 
    /* Check for overflow on the elements count (so the caller does not have to
     * check.)
@@ -143,7 +141,7 @@ png_realloc_array,(png_const_structrp png_ptr, png_const_voidp old_array,
    if (add_elements <= INT_MAX - old_elements)
    {
       png_voidp new_array = png_malloc_array_checked(png_ptr,
-          old_elements+add_elements, element_size);
+         old_elements+add_elements, element_size);
 
       if (new_array != NULL)
       {
@@ -154,12 +152,21 @@ png_realloc_array,(png_const_structrp png_ptr, png_const_voidp old_array,
             memcpy(new_array, old_array, element_size*(unsigned)old_elements);
 
          memset((char*)new_array + element_size*(unsigned)old_elements, 0,
-             element_size*(unsigned)add_elements);
+            element_size*(unsigned)add_elements);
 
          return new_array;
       }
    }
 
+#ifdef PNG_READ_SUPPORTED
+# ifdef PNG_USER_LIMITS_SUPPORTED
+   /* The potential overflow case.  Set the cache counter so libpng will
+    * not make any more attempts
+    */
+   png_ptr->user_chunk_cache_max = 2;
+# endif
+#endif
+
    return NULL; /* error */
 }
 #endif /* TEXT || sPLT || STORE_UNKNOWN_CHUNKS */
@@ -179,38 +186,18 @@ png_malloc,(png_const_structrp png_ptr, png_alloc_size_t size),PNG_ALLOCATED)
    ret = png_malloc_base(png_ptr, size);
 
    if (ret == NULL)
-       png_error(png_ptr, "Out of memory"); /* 'm' means png_malloc */
+       png_error(png_ptr, "Out of memory");
 
    return ret;
 }
 
-#ifdef PNG_USER_MEM_SUPPORTED
-PNG_FUNCTION(png_voidp,PNGAPI
-png_malloc_default,(png_const_structrp png_ptr, png_alloc_size_t size),
-    PNG_ALLOCATED PNG_DEPRECATED)
-{
-   png_voidp ret;
-
-   if (png_ptr == NULL)
-      return NULL;
-
-   /* Passing 'NULL' here bypasses the application provided memory handler. */
-   ret = png_malloc_base(NULL/*use malloc*/, size);
-
-   if (ret == NULL)
-      png_error(png_ptr, "Out of Memory"); /* 'M' means png_malloc_default */
-
-   return ret;
-}
-#endif /* USER_MEM */
-
 /* This function was added at libpng version 1.2.3.  The png_malloc_warn()
  * function will issue a png_warning and return NULL instead of issuing a
  * png_error, if it fails to allocate the requested memory.
  */
 PNG_FUNCTION(png_voidp,PNGAPI
 png_malloc_warn,(png_const_structrp png_ptr, png_alloc_size_t size),
-    PNG_ALLOCATED)
+   PNG_ALLOCATED)
 {
    if (png_ptr != NULL)
    {
@@ -239,17 +226,8 @@ png_free(png_const_structrp png_ptr, png_voidp ptr)
       png_ptr->free_fn(png_constcast(png_structrp,png_ptr), ptr);
 
    else
-      png_free_default(png_ptr, ptr);
-}
-
-PNG_FUNCTION(void,PNGAPI
-png_free_default,(png_const_structrp png_ptr, png_voidp ptr),PNG_DEPRECATED)
-{
-   if (png_ptr == NULL || ptr == NULL)
-      return;
 #endif /* USER_MEM */
-
-   free(ptr);
+      free(ptr);
 }
 
 #ifdef PNG_USER_MEM_SUPPORTED