diff options
Diffstat (limited to 'src/3rdparty/libjpeg/src')
63 files changed, 1835 insertions, 1058 deletions
diff --git a/src/3rdparty/libjpeg/src/ChangeLog.md b/src/3rdparty/libjpeg/src/ChangeLog.md index 59fb2de96a..a547522a15 100644 --- a/src/3rdparty/libjpeg/src/ChangeLog.md +++ b/src/3rdparty/libjpeg/src/ChangeLog.md @@ -1,3 +1,408 @@ +2.1.5 +===== + +### Significant changes relative to 2.1.4: + +1. Fixed issues in the build system whereby, when using the Ninja Multi-Config +CMake generator, a static build of libjpeg-turbo (a build in which +`ENABLE_SHARED` is `0`) could not be installed, a Windows installer could not +be built, and the Java regression tests failed. + +2. Fixed a regression introduced by 2.0 beta1[15] that caused a buffer overrun +in the progressive Huffman encoder when attempting to transform a +specially-crafted malformed 12-bit-per-component JPEG image into a progressive +12-bit-per-component JPEG image using a 12-bit-per-component build of +libjpeg-turbo (`-DWITH_12BIT=1`.) Given that the buffer overrun was fully +contained within the progressive Huffman encoder structure and did not cause a +segfault or other user-visible errant behavior, given that the lossless +transformer (unlike the decompressor) is not generally exposed to arbitrary +data exploits, and given that 12-bit-per-component builds of libjpeg-turbo are +uncommon, this issue did not likely pose a security risk. + +3. Fixed an issue whereby, when using a 12-bit-per-component build of +libjpeg-turbo (`-DWITH_12BIT=1`), passing samples with values greater than 4095 +or less than 0 to `jpeg_write_scanlines()` caused a buffer overrun or underrun +in the RGB-to-YCbCr color converter. + +4. Fixed a floating point exception that occurred when attempting to use the +jpegtran `-drop` and `-trim` options to losslessly transform a +specially-crafted malformed JPEG image. + +5. Fixed an issue in `tjBufSizeYUV2()` whereby it returned a bogus result, +rather than throwing an error, if the `align` parameter was not a power of 2. +Fixed a similar issue in `tjCompressFromYUV()` whereby it generated a corrupt +JPEG image in certain cases, rather than throwing an error, if the `align` +parameter was not a power of 2. + +6. Fixed an issue whereby `tjDecompressToYUV2()`, which is a wrapper for +`tjDecompressToYUVPlanes()`, used the desired YUV image dimensions rather than +the actual scaled image dimensions when computing the plane pointers and +strides to pass to `tjDecompressToYUVPlanes()`. This caused a buffer overrun +and subsequent segfault if the desired image dimensions exceeded the scaled +image dimensions. + +7. Fixed an issue whereby, when decompressing a 12-bit-per-component JPEG image +(`-DWITH_12BIT=1`) using an alpha-enabled output color space such as +`JCS_EXT_RGBA`, the alpha channel was set to 255 rather than 4095. + +8. Fixed an issue whereby the Java version of TJBench did not accept a range of +quality values. + +9. Fixed an issue whereby, when `-progressive` was passed to TJBench, the JPEG +input image was not transformed into a progressive JPEG image prior to +decompression. + + +2.1.4 +===== + +### Significant changes relative to 2.1.3: + +1. Fixed a regression introduced in 2.1.3 that caused build failures with +Visual Studio 2010. + +2. The `tjDecompressHeader3()` function in the TurboJPEG C API and the +`TJDecompressor.setSourceImage()` method in the TurboJPEG Java API now accept +"abbreviated table specification" (AKA "tables-only") datastreams, which can be +used to prime the decompressor with quantization and Huffman tables that can be +used when decompressing subsequent "abbreviated image" datastreams. + +3. libjpeg-turbo now performs run-time detection of AltiVec instructions on +OS X/PowerPC systems if AltiVec instructions are not enabled at compile time. +This allows both AltiVec-equipped (PowerPC G4 and G5) and non-AltiVec-equipped +(PowerPC G3) CPUs to be supported using the same build of libjpeg-turbo. + +4. Fixed an error ("Bogus virtual array access") that occurred when attempting +to decompress a progressive JPEG image with a height less than or equal to one +iMCU (8 * the vertical sampling factor) using buffered-image mode with +interblock smoothing enabled. This was a regression introduced by +2.1 beta1[6(b)]. + +5. Fixed two issues that prevented partial image decompression from working +properly with buffered-image mode: + + - Attempting to call `jpeg_crop_scanline()` after +`jpeg_start_decompress()` but before `jpeg_start_output()` resulted in an error +("Improper call to JPEG library in state 207".) + - Attempting to use `jpeg_skip_scanlines()` resulted in an error ("Bogus +virtual array access") under certain circumstances. + + +2.1.3 +===== + +### Significant changes relative to 2.1.2: + +1. Fixed a regression introduced by 2.0 beta1[7] whereby cjpeg compressed PGM +input files into full-color JPEG images unless the `-grayscale` option was +used. + +2. cjpeg now automatically compresses GIF and 8-bit BMP input files into +grayscale JPEG images if the input files contain only shades of gray. + +3. The build system now enables the intrinsics implementation of the AArch64 +(Arm 64-bit) Neon SIMD extensions by default when using GCC 12 or later. + +4. Fixed a segfault that occurred while decompressing a 4:2:0 JPEG image using +the merged (non-fancy) upsampling algorithms (that is, with +`cinfo.do_fancy_upsampling` set to `FALSE`) along with `jpeg_crop_scanline()`. +Specifically, the segfault occurred if the number of bytes remaining in the +output buffer was less than the number of bytes required to represent one +uncropped scanline of the output image. For that reason, the issue could only +be reproduced using the libjpeg API, not using djpeg. + + +2.1.2 +===== + +### Significant changes relative to 2.1.1: + +1. Fixed a regression introduced by 2.1 beta1[13] that caused the remaining +GAS implementations of AArch64 (Arm 64-bit) Neon SIMD functions (which are used +by default with GCC for performance reasons) to be placed in the `.rodata` +section rather than in the `.text` section. This caused the GNU linker to +automatically place the `.rodata` section in an executable segment, which +prevented libjpeg-turbo from working properly with other linkers and also +represented a potential security risk. + +2. Fixed an issue whereby the `tjTransform()` function incorrectly computed the +MCU block size for 4:4:4 JPEG images with non-unary sampling factors and thus +unduly rejected some cropping regions, even though those regions aligned with +8x8 MCU block boundaries. + +3. Fixed a regression introduced by 2.1 beta1[13] that caused the build system +to enable the Arm Neon SIMD extensions when targetting Armv6 and other legacy +architectures that do not support Neon instructions. + +4. libjpeg-turbo now performs run-time detection of AltiVec instructions on +FreeBSD/PowerPC systems if AltiVec instructions are not enabled at compile +time. This allows both AltiVec-equipped and non-AltiVec-equipped CPUs to be +supported using the same build of libjpeg-turbo. + +5. cjpeg now accepts a `-strict` argument similar to that of djpeg and +jpegtran, which causes the compressor to abort if an LZW-compressed GIF input +image contains incomplete or corrupt image data. + + +2.1.1 +===== + +### Significant changes relative to 2.1.0: + +1. Fixed a regression introduced in 2.1.0 that caused build failures with +non-GCC-compatible compilers for Un*x/Arm platforms. + +2. Fixed a regression introduced by 2.1 beta1[13] that prevented the Arm 32-bit +(AArch32) Neon SIMD extensions from building unless the C compiler flags +included `-mfloat-abi=softfp` or `-mfloat-abi=hard`. + +3. Fixed an issue in the AArch32 Neon SIMD Huffman encoder whereby reliance on +undefined C compiler behavior led to crashes ("SIGBUS: illegal alignment") on +Android systems when running AArch32/Thumb builds of libjpeg-turbo built with +recent versions of Clang. + +4. Added a command-line argument (`-copy icc`) to jpegtran that causes it to +copy only the ICC profile markers from the source file and discard any other +metadata. + +5. libjpeg-turbo should now build and run on CHERI-enabled architectures, which +use capability pointers that are larger than the size of `size_t`. + +6. Fixed a regression (CVE-2021-37972) introduced by 2.1 beta1[5] that caused a +segfault in the 64-bit SSE2 Huffman encoder when attempting to losslessly +transform a specially-crafted malformed JPEG image. + + +2.1.0 +===== + +### Significant changes relative to 2.1 beta1: + +1. Fixed a regression introduced by 2.1 beta1[6(b)] whereby attempting to +decompress certain progressive JPEG images with one or more component planes of +width 8 or less caused a buffer overrun. + +2. Fixed a regression introduced by 2.1 beta1[6(b)] whereby attempting to +decompress a specially-crafted malformed progressive JPEG image caused the +block smoothing algorithm to read from uninitialized memory. + +3. Fixed an issue in the Arm Neon SIMD Huffman encoders that caused the +encoders to generate incorrect results when using the Clang compiler with +Visual Studio. + +4. Fixed a floating point exception (CVE-2021-20205) that occurred when +attempting to compress a specially-crafted malformed GIF image with a specified +image width of 0 using cjpeg. + +5. Fixed a regression introduced by 2.0 beta1[15] whereby attempting to +generate a progressive JPEG image on an SSE2-capable CPU using a scan script +containing one or more scans with lengths divisible by 32 and non-zero +successive approximation low bit positions would, under certain circumstances, +result in an error ("Missing Huffman code table entry") and an invalid JPEG +image. + +6. Introduced a new flag (`TJFLAG_LIMITSCANS` in the TurboJPEG C API and +`TJ.FLAG_LIMIT_SCANS` in the TurboJPEG Java API) and a corresponding TJBench +command-line argument (`-limitscans`) that causes the TurboJPEG decompression +and transform functions/operations to return/throw an error if a progressive +JPEG image contains an unreasonably large number of scans. This allows +applications that use the TurboJPEG API to guard against an exploit of the +progressive JPEG format described in the report +["Two Issues with the JPEG Standard"](https://libjpeg-turbo.org/pmwiki/uploads/About/TwoIssueswiththeJPEGStandard.pdf). + +7. The PPM reader now throws an error, rather than segfaulting (due to a buffer +overrun, CVE-2021-46822) or generating incorrect pixels, if an application +attempts to use the `tjLoadImage()` function to load a 16-bit binary PPM file +(a binary PPM file with a maximum value greater than 255) into a grayscale +image buffer or to load a 16-bit binary PGM file into an RGB image buffer. + +8. Fixed an issue in the PPM reader that caused incorrect pixels to be +generated when using the `tjLoadImage()` function to load a 16-bit binary PPM +file into an extended RGB image buffer. + +9. Fixed an issue whereby, if a JPEG buffer was automatically re-allocated by +one of the TurboJPEG compression or transform functions and an error +subsequently occurred during compression or transformation, the JPEG buffer +pointer passed by the application was not updated when the function returned. + + +2.0.90 (2.1 beta1) +================== + +### Significant changes relative to 2.0.6: + +1. The build system, x86-64 SIMD extensions, and accelerated Huffman codec now +support the x32 ABI on Linux, which allows for using x86-64 instructions with +32-bit pointers. The x32 ABI is generally enabled by adding `-mx32` to the +compiler flags. + + Caveats: + - CMake 3.9.0 or later is required in order for the build system to +automatically detect an x32 build. + - Java does not support the x32 ABI, and thus the TurboJPEG Java API will +automatically be disabled with x32 builds. + +2. Added Loongson MMI SIMD implementations of the RGB-to-grayscale, 4:2:2 fancy +chroma upsampling, 4:2:2 and 4:2:0 merged chroma upsampling/color conversion, +and fast integer DCT/IDCT algorithms. Relative to libjpeg-turbo 2.0.x, this +speeds up: + + - the compression of RGB source images into grayscale JPEG images by +approximately 20% + - the decompression of 4:2:2 JPEG images by approximately 40-60% when +using fancy upsampling + - the decompression of 4:2:2 and 4:2:0 JPEG images by approximately +15-20% when using merged upsampling + - the compression of RGB source images by approximately 30-45% when using +the fast integer DCT + - the decompression of JPEG images into RGB destination images by +approximately 2x when using the fast integer IDCT + + The overall decompression speedup for RGB images is now approximately +2.3-3.7x (compared to 2-3.5x with libjpeg-turbo 2.0.x.) + +3. 32-bit (Armv7 or Armv7s) iOS builds of libjpeg-turbo are no longer +supported, and the libjpeg-turbo build system can no longer be used to package +such builds. 32-bit iOS apps cannot run in iOS 11 and later, and the App Store +no longer allows them. + +4. 32-bit (i386) OS X/macOS builds of libjpeg-turbo are no longer supported, +and the libjpeg-turbo build system can no longer be used to package such +builds. 32-bit Mac applications cannot run in macOS 10.15 "Catalina" and +later, and the App Store no longer allows them. + +5. The SSE2 (x86 SIMD) and C Huffman encoding algorithms have been +significantly optimized, resulting in a measured average overall compression +speedup of 12-28% for 64-bit code and 22-52% for 32-bit code on various Intel +and AMD CPUs, as well as a measured average overall compression speedup of +0-23% on platforms that do not have a SIMD-accelerated Huffman encoding +implementation. + +6. The block smoothing algorithm that is applied by default when decompressing +progressive Huffman-encoded JPEG images has been improved in the following +ways: + + - The algorithm is now more fault-tolerant. Previously, if a particular +scan was incomplete, then the smoothing parameters for the incomplete scan +would be applied to the entire output image, including the parts of the image +that were generated by the prior (complete) scan. Visually, this had the +effect of removing block smoothing from lower-frequency scans if they were +followed by an incomplete higher-frequency scan. libjpeg-turbo now applies +block smoothing parameters to each iMCU row based on which scan generated the +pixels in that row, rather than always using the block smoothing parameters for +the most recent scan. + - When applying block smoothing to DC scans, a Gaussian-like kernel with a +5x5 window is used to reduce the "blocky" appearance. + +7. Added SIMD acceleration for progressive Huffman encoding on Arm platforms. +This speeds up the compression of full-color progressive JPEGs by about 30-40% +on average (relative to libjpeg-turbo 2.0.x) when using modern Arm CPUs. + +8. Added configure-time and run-time auto-detection of Loongson MMI SIMD +instructions, so that the Loongson MMI SIMD extensions can be included in any +MIPS64 libjpeg-turbo build. + +9. Added fault tolerance features to djpeg and jpegtran, mainly to demonstrate +methods by which applications can guard against the exploits of the JPEG format +described in the report +["Two Issues with the JPEG Standard"](https://libjpeg-turbo.org/pmwiki/uploads/About/TwoIssueswiththeJPEGStandard.pdf). + + - Both programs now accept a `-maxscans` argument, which can be used to +limit the number of allowable scans in the input file. + - Both programs now accept a `-strict` argument, which can be used to +treat all warnings as fatal. + +10. CMake package config files are now included for both the libjpeg and +TurboJPEG API libraries. This facilitates using libjpeg-turbo with CMake's +`find_package()` function. For example: + + find_package(libjpeg-turbo CONFIG REQUIRED) + + add_executable(libjpeg_program libjpeg_program.c) + target_link_libraries(libjpeg_program PUBLIC libjpeg-turbo::jpeg) + + add_executable(libjpeg_program_static libjpeg_program.c) + target_link_libraries(libjpeg_program_static PUBLIC + libjpeg-turbo::jpeg-static) + + add_executable(turbojpeg_program turbojpeg_program.c) + target_link_libraries(turbojpeg_program PUBLIC + libjpeg-turbo::turbojpeg) + + add_executable(turbojpeg_program_static turbojpeg_program.c) + target_link_libraries(turbojpeg_program_static PUBLIC + libjpeg-turbo::turbojpeg-static) + +11. Since the Unisys LZW patent has long expired, cjpeg and djpeg can now +read/write both LZW-compressed and uncompressed GIF files (feature ported from +jpeg-6a and jpeg-9d.) + +12. jpegtran now includes the `-wipe` and `-drop` options from jpeg-9a and +jpeg-9d, as well as the ability to expand the image size using the `-crop` +option. Refer to jpegtran.1 or usage.txt for more details. + +13. Added a complete intrinsics implementation of the Arm Neon SIMD extensions, +thus providing SIMD acceleration on Arm platforms for all of the algorithms +that are SIMD-accelerated on x86 platforms. This new implementation is +significantly faster in some cases than the old GAS implementation-- +depending on the algorithms used, the type of CPU core, and the compiler. GCC, +as of this writing, does not provide a full or optimal set of Neon intrinsics, +so for performance reasons, the default when building libjpeg-turbo with GCC is +to continue using the GAS implementation of the following algorithms: + + - 32-bit RGB-to-YCbCr color conversion + - 32-bit fast and accurate inverse DCT + - 64-bit RGB-to-YCbCr and YCbCr-to-RGB color conversion + - 64-bit accurate forward and inverse DCT + - 64-bit Huffman encoding + + A new CMake variable (`NEON_INTRINSICS`) can be used to override this +default. + + Since the new intrinsics implementation includes SIMD acceleration +for merged upsampling/color conversion, 1.5.1[5] is no longer necessary and has +been reverted. + +14. The Arm Neon SIMD extensions can now be built using Visual Studio. + +15. The build system can now be used to generate a universal x86-64 + Armv8 +libjpeg-turbo SDK package for both iOS and macOS. + + +2.0.6 +===== + +### Significant changes relative to 2.0.5: + +1. Fixed "using JNI after critical get" errors that occurred on Android +platforms when using any of the YUV encoding/compression/decompression/decoding +methods in the TurboJPEG Java API. + +2. Fixed or worked around multiple issues with `jpeg_skip_scanlines()`: + + - Fixed segfaults (CVE-2020-35538) or "Corrupt JPEG data: premature end of +data segment" errors in `jpeg_skip_scanlines()` that occurred when +decompressing 4:2:2 or 4:2:0 JPEG images using merged (non-fancy) +upsampling/color conversion (that is, when setting `cinfo.do_fancy_upsampling` +to `FALSE`.) 2.0.0[6] was a similar fix, but it did not cover all cases. + - `jpeg_skip_scanlines()` now throws an error if two-pass color +quantization is enabled. Two-pass color quantization never worked properly +with `jpeg_skip_scanlines()`, and the issues could not readily be fixed. + - Fixed an issue whereby `jpeg_skip_scanlines()` always returned 0 when +skipping past the end of an image. + +3. The Arm 64-bit (Armv8) Neon SIMD extensions can now be built using MinGW +toolchains targetting Arm64 (AArch64) Windows binaries. + +4. Fixed unexpected visual artifacts that occurred when using +`jpeg_crop_scanline()` and interblock smoothing while decompressing only the DC +scan of a progressive JPEG image. + +5. Fixed an issue whereby libjpeg-turbo would not build if 12-bit-per-component +JPEG support (`WITH_12BIT`) was enabled along with libjpeg v7 or libjpeg v8 +API/ABI emulation (`WITH_JPEG7` or `WITH_JPEG8`.) + + 2.0.5 ===== @@ -54,17 +459,17 @@ JPEG images. This was known to cause a buffer overflow when attempting to decompress some such images using `tjDecompressToYUV2()` or `tjDecompressToYUVPlanes()`. -5. Fixed an issue, detected by ASan, whereby attempting to losslessly transform -a specially-crafted malformed JPEG image containing an extremely-high-frequency -coefficient block (junk image data that could never be generated by a -legitimate JPEG compressor) could cause the Huffman encoder's local buffer to -be overrun. (Refer to 1.4.0[9] and 1.4beta1[15].) Given that the buffer -overrun was fully contained within the stack and did not cause a segfault or -other user-visible errant behavior, and given that the lossless transformer -(unlike the decompressor) is not generally exposed to arbitrary data exploits, -this issue did not likely pose a security risk. - -6. The ARM 64-bit (ARMv8) NEON SIMD assembly code now stores constants in a +5. Fixed an issue (CVE-2020-17541), detected by ASan, whereby attempting to +losslessly transform a specially-crafted malformed JPEG image containing an +extremely-high-frequency coefficient block (junk image data that could never be +generated by a legitimate JPEG compressor) could cause the Huffman encoder's +local buffer to be overrun. (Refer to 1.4.0[9] and 1.4beta1[15].) Given that +the buffer overrun was fully contained within the stack and did not cause a +segfault or other user-visible errant behavior, and given that the lossless +transformer (unlike the decompressor) is not generally exposed to arbitrary +data exploits, this issue did not likely pose a security risk. + +6. The Arm 64-bit (Armv8) Neon SIMD assembly code now stores constants in a separate read-only data section rather than in the text section, to support execute-only memory layouts. @@ -246,7 +651,7 @@ detect actual security issues, should they arise in the future. 1. Added AVX2 SIMD implementations of the colorspace conversion, chroma downsampling and upsampling, integer quantization and sample conversion, and -slow integer DCT/IDCT algorithms. When using the slow integer DCT/IDCT +accurate integer DCT/IDCT algorithms. When using the accurate integer DCT/IDCT algorithms on AVX2-equipped CPUs, the compression of RGB images is approximately 13-36% (avg. 22%) faster (relative to libjpeg-turbo 1.5.x) with 64-bit code and 11-21% (avg. 17%) faster with 32-bit code, and the @@ -350,16 +755,16 @@ algorithm that caused incorrect dithering in the output image. This algorithm now produces bitwise-identical results to the unmerged algorithms. 12. The SIMD function symbols for x86[-64]/ELF, MIPS/ELF, macOS/x86[-64] (if -libjpeg-turbo is built with YASM), and iOS/ARM[64] builds are now private. +libjpeg-turbo is built with Yasm), and iOS/Arm[64] builds are now private. This prevents those symbols from being exposed in applications or shared libraries that link statically with libjpeg-turbo. 13. Added Loongson MMI SIMD implementations of the RGB-to-YCbCr and YCbCr-to-RGB colorspace conversion, 4:2:0 chroma downsampling, 4:2:0 fancy -chroma upsampling, integer quantization, and slow integer DCT/IDCT algorithms. -When using the slow integer DCT/IDCT, this speeds up the compression of RGB -images by approximately 70-100% and the decompression of RGB images by -approximately 2-3.5x. +chroma upsampling, integer quantization, and accurate integer DCT/IDCT +algorithms. When using the accurate integer DCT/IDCT, this speeds up the +compression of RGB images by approximately 70-100% and the decompression of RGB +images by approximately 2-3.5x. 14. Fixed a build error when building with older MinGW releases (regression caused by 1.5.1[7].) @@ -409,9 +814,9 @@ end of a single-scan (non-progressive) image, subsequent calls to `jpeg_consume_input()` would return `JPEG_SUSPENDED` rather than `JPEG_REACHED_EOI`. -9. `jpeg_crop_scanlines()` now works correctly when decompressing grayscale -JPEG images that were compressed with a sampling factor other than 1 (for -instance, with `cjpeg -grayscale -sample 2x2`). +9. `jpeg_crop_scanline()` now works correctly when decompressing grayscale JPEG +images that were compressed with a sampling factor other than 1 (for instance, +with `cjpeg -grayscale -sample 2x2`). 1.5.2 @@ -435,7 +840,7 @@ on PowerPC-based AmigaOS 4 and OpenBSD systems. 5. Fixed build and runtime errors on Windows that occurred when building libjpeg-turbo with libjpeg v7 API/ABI emulation and the in-memory source/destination managers. Due to an oversight, the `jpeg_skip_scanlines()` -and `jpeg_crop_scanlines()` functions were not being included in jpeg7.dll when +and `jpeg_crop_scanline()` functions were not being included in jpeg7.dll when libjpeg-turbo was built with `-DWITH_JPEG7=1` and `-DWITH_MEMSRCDST=1`. 6. Fixed "Bogus virtual array access" error that occurred when using the @@ -691,8 +1096,8 @@ benchmarking or regression testing, SIMD-accelerated Huffman encoding can be disabled by setting the `JSIMD_NOHUFFENC` environment variable to `1`. 13. Added ARM 64-bit (ARMv8) NEON SIMD implementations of the commonly-used -compression algorithms (including the slow integer forward DCT and h2v2 & h2v1 -downsampling algorithms, which are not accelerated in the 32-bit NEON +compression algorithms (including the accurate integer forward DCT and h2v2 & +h2v1 downsampling algorithms, which are not accelerated in the 32-bit NEON implementation.) This speeds up the compression of full-color JPEGs by about 75% on average on a Cavium ThunderX processor and by about 2-2.5x on average on Cortex-A53 and Cortex-A57 cores. @@ -823,8 +1228,8 @@ platforms other than Windows or Linux. Oops. 7. Fixed an extremely rare bug in the Huffman encoder that caused 64-bit builds of libjpeg-turbo to incorrectly encode a few specific test images when -quality=98, an optimized Huffman table, and the slow integer forward DCT were -used. +quality=98, an optimized Huffman table, and the accurate integer forward DCT +were used. 8. The Windows (CMake) build system now supports building only static or only shared libraries. This is accomplished by adding either `-DENABLE_STATIC=0` or @@ -983,8 +1388,8 @@ floating point inverse DCT (using code borrowed from libjpeg v8a and later.) The accuracy of this implementation now matches the accuracy of the SSE/SSE2 implementation. Note, however, that the floating point DCT/IDCT algorithms are mainly a legacy feature. They generally do not produce significantly better -accuracy than the slow integer DCT/IDCT algorithms, and they are quite a bit -slower. +accuracy than the accurate integer DCT/IDCT algorithms, and they are quite a +bit slower. 8. Added a new output colorspace (`JCS_RGB565`) to the libjpeg API that allows for decompressing JPEG images into RGB565 (16-bit) pixels. If dithering is not @@ -1235,8 +1640,8 @@ either the fast or the accurate DCT/IDCT algorithms in the underlying codec. ### Significant changes relative to 1.2 beta1: -1. Fixed build issue with YASM on Unix systems (the libjpeg-turbo build system -was not adding the current directory to the assembler include path, so YASM +1. Fixed build issue with Yasm on Unix systems (the libjpeg-turbo build system +was not adding the current directory to the assembler include path, so Yasm was not able to find jsimdcfg.inc.) 2. Fixed out-of-bounds read in SSE2 SIMD code that occurred when decompressing @@ -1304,7 +1709,7 @@ transposed or rotated 90 degrees. 8. All legacy VirtualGL code has been re-factored, and this has allowed libjpeg-turbo, in its entirety, to be re-licensed under a BSD-style license. -9. libjpeg-turbo can now be built with YASM. +9. libjpeg-turbo can now be built with Yasm. 10. Added SIMD acceleration for ARM Linux and iOS platforms that support NEON instructions. @@ -1394,8 +1799,8 @@ cases. 2. Despite the above, the fast integer forward DCT still degrades somewhat for JPEG qualities greater than 95, so the TurboJPEG wrapper will now automatically -use the slow integer forward DCT when generating JPEG images of quality 96 or -greater. This reduces compression performance by as much as 15% for these +use the accurate integer forward DCT when generating JPEG images of quality 96 +or greater. This reduces compression performance by as much as 15% for these high-quality images but is necessary to ensure that the images are perceptually lossless. It also ensures that the library can avoid the performance pitfall created by [1]. diff --git a/src/3rdparty/libjpeg/src/README.ijg b/src/3rdparty/libjpeg/src/README.ijg index 2e39f965c2..9453c19501 100644 --- a/src/3rdparty/libjpeg/src/README.ijg +++ b/src/3rdparty/libjpeg/src/README.ijg @@ -128,7 +128,7 @@ with respect to this software, its quality, accuracy, merchantability, or fitness for a particular purpose. This software is provided "AS IS", and you, its user, assume the entire risk as to its quality and accuracy. -This software is copyright (C) 1991-2016, Thomas G. Lane, Guido Vollbeding. +This software is copyright (C) 1991-2020, Thomas G. Lane, Guido Vollbeding. All Rights Reserved except as specified below. Permission is hereby granted to use, copy, modify, and distribute this @@ -159,19 +159,6 @@ commercial products, provided that all warranty or liability claims are assumed by the product vendor. -The IJG distribution formerly included code to read and write GIF files. -To avoid entanglement with the Unisys LZW patent (now expired), GIF reading -support has been removed altogether, and the GIF writer has been simplified -to produce "uncompressed GIFs". This technique does not use the LZW -algorithm; the resulting GIF files are larger than usual, but are readable -by all standard GIF decoders. - -We are required to state that - "The Graphics Interchange Format(c) is the Copyright property of - CompuServe Incorporated. GIF(sm) is a Service Mark property of - CompuServe Incorporated." - - REFERENCES ========== @@ -223,12 +210,12 @@ https://www.iso.org/standard/54989.html and http://www.itu.int/rec/T-REC-T.871. A PDF file of the older JFIF 1.02 specification is available at http://www.w3.org/Graphics/JPEG/jfif3.pdf. -The TIFF 6.0 file format specification can be obtained by FTP from -ftp://ftp.sgi.com/graphics/tiff/TIFF6.ps.gz. The JPEG incorporation scheme -found in the TIFF 6.0 spec of 3-June-92 has a number of serious problems. -IJG does not recommend use of the TIFF 6.0 design (TIFF Compression tag 6). -Instead, we recommend the JPEG design proposed by TIFF Technical Note #2 -(Compression tag 7). Copies of this Note can be obtained from +The TIFF 6.0 file format specification can be obtained from +http://mirrors.ctan.org/graphics/tiff/TIFF6.ps.gz. The JPEG incorporation +scheme found in the TIFF 6.0 spec of 3-June-92 has a number of serious +problems. IJG does not recommend use of the TIFF 6.0 design (TIFF Compression +tag 6). Instead, we recommend the JPEG design proposed by TIFF Technical Note +#2 (Compression tag 7). Copies of this Note can be obtained from http://www.ijg.org/files/. It is expected that the next revision of the TIFF spec will replace the 6.0 JPEG design with the Note's design. Although IJG's own code does not support TIFF/JPEG, the free libtiff library @@ -243,14 +230,8 @@ The most recent released version can always be found there in directory "files". The JPEG FAQ (Frequently Asked Questions) article is a source of some -general information about JPEG. -It is available on the World Wide Web at http://www.faqs.org/faqs/jpeg-faq/ -and other news.answers archive sites, including the official news.answers -archive at rtfm.mit.edu: ftp://rtfm.mit.edu/pub/usenet/news.answers/jpeg-faq/. -If you don't have Web or FTP access, send e-mail to mail-server@rtfm.mit.edu -with body - send usenet/news.answers/jpeg-faq/part1 - send usenet/news.answers/jpeg-faq/part2 +general information about JPEG. It is available at +http://www.faqs.org/faqs/jpeg-faq. FILE FORMAT COMPATIBILITY diff --git a/src/3rdparty/libjpeg/src/README.md b/src/3rdparty/libjpeg/src/README.md index e7ff743a47..01e391ea7c 100644 --- a/src/3rdparty/libjpeg/src/README.md +++ b/src/3rdparty/libjpeg/src/README.md @@ -2,8 +2,8 @@ Background ========== libjpeg-turbo is a JPEG image codec that uses SIMD instructions to accelerate -baseline JPEG compression and decompression on x86, x86-64, ARM, PowerPC, and -MIPS systems, as well as progressive JPEG compression on x86 and x86-64 +baseline JPEG compression and decompression on x86, x86-64, Arm, PowerPC, and +MIPS systems, as well as progressive JPEG compression on x86, x86-64, and Arm systems. On such systems, libjpeg-turbo is generally 2-6x as fast as libjpeg, all else being equal. On other types of systems, libjpeg-turbo can still outperform libjpeg by a significant amount, by virtue of its highly-optimized @@ -179,8 +179,8 @@ supported and which aren't. NOTE: As of this writing, extensive research has been conducted into the usefulness of DCT scaling as a means of data reduction and SmartScale as a -means of quality improvement. The reader is invited to peruse the research at -<http://www.libjpeg-turbo.org/About/SmartScale> and draw his/her own conclusions, +means of quality improvement. Readers are invited to peruse the research at +<http://www.libjpeg-turbo.org/About/SmartScale> and draw their own conclusions, but it is the general belief of our project that these features have not demonstrated sufficient usefulness to justify inclusion in libjpeg-turbo. @@ -287,12 +287,13 @@ following reasons: (and slightly faster) floating point IDCT algorithm introduced in libjpeg v8a as opposed to the algorithm used in libjpeg v6b. It should be noted, however, that this algorithm basically brings the accuracy of the floating - point IDCT in line with the accuracy of the slow integer IDCT. The floating - point DCT/IDCT algorithms are mainly a legacy feature, and they do not - produce significantly more accuracy than the slow integer algorithms (to put - numbers on this, the typical difference in PNSR between the two algorithms - is less than 0.10 dB, whereas changing the quality level by 1 in the upper - range of the quality scale is typically more like a 1.0 dB difference.) + point IDCT in line with the accuracy of the accurate integer IDCT. The + floating point DCT/IDCT algorithms are mainly a legacy feature, and they do + not produce significantly more accuracy than the accurate integer algorithms + (to put numbers on this, the typical difference in PNSR between the two + algorithms is less than 0.10 dB, whereas changing the quality level by 1 in + the upper range of the quality scale is typically more like a 1.0 dB + difference.) - If the floating point algorithms in libjpeg-turbo are not implemented using SIMD instructions on a particular platform, then the accuracy of the @@ -340,7 +341,7 @@ The algorithm used by the SIMD-accelerated quantization function cannot produce correct results whenever the fast integer forward DCT is used along with a JPEG quality of 98-100. Thus, libjpeg-turbo must use the non-SIMD quantization function in those cases. This causes performance to drop by as much as 40%. -It is therefore strongly advised that you use the slow integer forward DCT +It is therefore strongly advised that you use the accurate integer forward DCT whenever encoding images with a JPEG quality of 98 or higher. diff --git a/src/3rdparty/libjpeg/src/change.log b/src/3rdparty/libjpeg/src/change.log index f090d7788c..e4d0ddc4bd 100644 --- a/src/3rdparty/libjpeg/src/change.log +++ b/src/3rdparty/libjpeg/src/change.log @@ -6,6 +6,25 @@ reference. Please see ChangeLog.md for information specific to libjpeg-turbo. CHANGE LOG for Independent JPEG Group's JPEG software +Version 9d 12-Jan-2020 +----------------------- + +Restore GIF read and write support from libjpeg version 6a. +Thank to Wolfgang Werner (W.W.) Heinz for suggestion. + +Add jpegtran -drop option; add options to the crop extension and wipe +to fill the extra area with content from the source image region, +instead of gray out. + + +Version 9c 14-Jan-2018 +----------------------- + +jpegtran: add an option to the -wipe switch to fill the region +with the average of adjacent blocks, instead of gray out. +Thank to Caitlyn Feddock and Maddie Ziegler for inspiration. + + Version 9b 17-Jan-2016 ----------------------- @@ -13,6 +32,13 @@ Document 'f' specifier for jpegtran -crop specification. Thank to Michele Martone for suggestion. +Version 9a 19-Jan-2014 +----------------------- + +Add jpegtran -wipe option and extension for -crop. +Thank to Andrew Senior, David Clunie, and Josef Schmid for suggestion. + + Version 9 13-Jan-2013 ---------------------- @@ -138,11 +164,6 @@ Huffman tables being used. Huffman tables are checked for validity much more carefully than before. -To avoid the Unisys LZW patent, djpeg's GIF output capability has been -changed to produce "uncompressed GIFs", and cjpeg's GIF input capability -has been removed altogether. We're not happy about it either, but there -seems to be no good alternative. - The configure script now supports building libjpeg as a shared library on many flavors of Unix (all the ones that GNU libtool knows how to build shared libraries for). Use "./configure --enable-shared" to diff --git a/src/3rdparty/libjpeg/src/jcapimin.c b/src/3rdparty/libjpeg/src/jcapimin.c index 178c55ba47..84e7ecc9a7 100644 --- a/src/3rdparty/libjpeg/src/jcapimin.c +++ b/src/3rdparty/libjpeg/src/jcapimin.c @@ -4,8 +4,8 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1994-1998, Thomas G. Lane. * Modified 2003-2010 by Guido Vollbeding. - * It was modified by The libjpeg-turbo Project to include only code relevant - * to libjpeg-turbo. + * libjpeg-turbo Modifications: + * Copyright (C) 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -52,7 +52,7 @@ jpeg_CreateCompress(j_compress_ptr cinfo, int version, size_t structsize) { struct jpeg_error_mgr *err = cinfo->err; void *client_data = cinfo->client_data; /* ignore Purify complaint here */ - MEMZERO(cinfo, sizeof(struct jpeg_compress_struct)); + memset(cinfo, 0, sizeof(struct jpeg_compress_struct)); cinfo->err = err; cinfo->client_data = client_data; } diff --git a/src/3rdparty/libjpeg/src/jcarith.c b/src/3rdparty/libjpeg/src/jcarith.c index b6d093f70e..b1720521bf 100644 --- a/src/3rdparty/libjpeg/src/jcarith.c +++ b/src/3rdparty/libjpeg/src/jcarith.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Developed 1997-2009 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2015, 2018, D. R. Commander. + * Copyright (C) 2015, 2018, 2021-2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -338,14 +338,14 @@ emit_restart(j_compress_ptr cinfo, int restart_num) compptr = cinfo->cur_comp_info[ci]; /* DC needs no table for refinement scan */ if (cinfo->progressive_mode == 0 || (cinfo->Ss == 0 && cinfo->Ah == 0)) { - MEMZERO(entropy->dc_stats[compptr->dc_tbl_no], DC_STAT_BINS); + memset(entropy->dc_stats[compptr->dc_tbl_no], 0, DC_STAT_BINS); /* Reset DC predictions to 0 */ entropy->last_dc_val[ci] = 0; entropy->dc_context[ci] = 0; } /* AC needs no table when not present */ if (cinfo->progressive_mode == 0 || cinfo->Se) { - MEMZERO(entropy->ac_stats[compptr->ac_tbl_no], AC_STAT_BINS); + memset(entropy->ac_stats[compptr->ac_tbl_no], 0, AC_STAT_BINS); } } @@ -836,7 +836,7 @@ start_pass(j_compress_ptr cinfo, boolean gather_statistics) * We are fully adaptive here and need no extra * statistics gathering pass! */ - ERREXIT(cinfo, JERR_NOT_COMPILED); + ERREXIT(cinfo, JERR_NOTIMPL); /* We assume jcmaster.c already validated the progressive scan parameters. */ @@ -867,7 +867,7 @@ start_pass(j_compress_ptr cinfo, boolean gather_statistics) if (entropy->dc_stats[tbl] == NULL) entropy->dc_stats[tbl] = (unsigned char *)(*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, DC_STAT_BINS); - MEMZERO(entropy->dc_stats[tbl], DC_STAT_BINS); + memset(entropy->dc_stats[tbl], 0, DC_STAT_BINS); /* Initialize DC predictions to 0 */ entropy->last_dc_val[ci] = 0; entropy->dc_context[ci] = 0; @@ -880,7 +880,7 @@ start_pass(j_compress_ptr cinfo, boolean gather_statistics) if (entropy->ac_stats[tbl] == NULL) entropy->ac_stats[tbl] = (unsigned char *)(*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, AC_STAT_BINS); - MEMZERO(entropy->ac_stats[tbl], AC_STAT_BINS); + memset(entropy->ac_stats[tbl], 0, AC_STAT_BINS); #ifdef CALCULATE_SPECTRAL_CONDITIONING if (cinfo->progressive_mode) /* Section G.1.3.2: Set appropriate arithmetic conditioning value Kx */ diff --git a/src/3rdparty/libjpeg/src/jccolext.c b/src/3rdparty/libjpeg/src/jccolext.c index 19c955c9d6..20f891a633 100644 --- a/src/3rdparty/libjpeg/src/jccolext.c +++ b/src/3rdparty/libjpeg/src/jccolext.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1996, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2009-2012, 2015, D. R. Commander. + * Copyright (C) 2009-2012, 2015, 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -48,9 +48,9 @@ rgb_ycc_convert_internal(j_compress_ptr cinfo, JSAMPARRAY input_buf, outptr2 = output_buf[2][output_row]; output_row++; for (col = 0; col < num_cols; col++) { - r = GETJSAMPLE(inptr[RGB_RED]); - g = GETJSAMPLE(inptr[RGB_GREEN]); - b = GETJSAMPLE(inptr[RGB_BLUE]); + r = RANGE_LIMIT(inptr[RGB_RED]); + g = RANGE_LIMIT(inptr[RGB_GREEN]); + b = RANGE_LIMIT(inptr[RGB_BLUE]); inptr += RGB_PIXELSIZE; /* If the inputs are 0..MAXJSAMPLE, the outputs of these equations * must be too; we do not need an explicit range-limiting operation. @@ -100,9 +100,9 @@ rgb_gray_convert_internal(j_compress_ptr cinfo, JSAMPARRAY input_buf, outptr = output_buf[0][output_row]; output_row++; for (col = 0; col < num_cols; col++) { - r = GETJSAMPLE(inptr[RGB_RED]); - g = GETJSAMPLE(inptr[RGB_GREEN]); - b = GETJSAMPLE(inptr[RGB_BLUE]); + r = RANGE_LIMIT(inptr[RGB_RED]); + g = RANGE_LIMIT(inptr[RGB_GREEN]); + b = RANGE_LIMIT(inptr[RGB_BLUE]); inptr += RGB_PIXELSIZE; /* Y */ outptr[col] = (JSAMPLE)((ctab[r + R_Y_OFF] + ctab[g + G_Y_OFF] + @@ -135,9 +135,9 @@ rgb_rgb_convert_internal(j_compress_ptr cinfo, JSAMPARRAY input_buf, outptr2 = output_buf[2][output_row]; output_row++; for (col = 0; col < num_cols; col++) { - outptr0[col] = GETJSAMPLE(inptr[RGB_RED]); - outptr1[col] = GETJSAMPLE(inptr[RGB_GREEN]); - outptr2[col] = GETJSAMPLE(inptr[RGB_BLUE]); + outptr0[col] = inptr[RGB_RED]; + outptr1[col] = inptr[RGB_GREEN]; + outptr2[col] = inptr[RGB_BLUE]; inptr += RGB_PIXELSIZE; } } diff --git a/src/3rdparty/libjpeg/src/jccolor.c b/src/3rdparty/libjpeg/src/jccolor.c index 036f6016d1..fb9f1cc192 100644 --- a/src/3rdparty/libjpeg/src/jccolor.c +++ b/src/3rdparty/libjpeg/src/jccolor.c @@ -5,7 +5,7 @@ * Copyright (C) 1991-1996, Thomas G. Lane. * libjpeg-turbo Modifications: * Copyright 2009 Pierre Ossman <ossman@cendio.se> for Cendio AB - * Copyright (C) 2009-2012, 2015, D. R. Commander. + * Copyright (C) 2009-2012, 2015, 2022, D. R. Commander. * Copyright (C) 2014, MIPS Technologies, Inc., California. * For conditions of distribution and use, see the accompanying README.ijg * file. @@ -17,7 +17,6 @@ #include "jinclude.h" #include "jpeglib.h" #include "jsimd.h" -#include "jconfigint.h" /* Private subobject */ @@ -84,6 +83,18 @@ typedef my_color_converter *my_cconvert_ptr; #define B_CR_OFF (7 * (MAXJSAMPLE + 1)) #define TABLE_SIZE (8 * (MAXJSAMPLE + 1)) +/* 12-bit samples use a 16-bit data type, so it is possible to pass + * out-of-range sample values (< 0 or > 4095) to jpeg_write_scanlines(). + * Thus, we mask the incoming 12-bit samples to guard against overrunning + * or underrunning the conversion tables. + */ + +#if BITS_IN_JSAMPLE == 12 +#define RANGE_LIMIT(value) ((value) & 0xFFF) +#else +#define RANGE_LIMIT(value) (value) +#endif + /* Include inline routines for colorspace extensions */ @@ -392,11 +403,11 @@ cmyk_ycck_convert(j_compress_ptr cinfo, JSAMPARRAY input_buf, outptr3 = output_buf[3][output_row]; output_row++; for (col = 0; col < num_cols; col++) { - r = MAXJSAMPLE - GETJSAMPLE(inptr[0]); - g = MAXJSAMPLE - GETJSAMPLE(inptr[1]); - b = MAXJSAMPLE - GETJSAMPLE(inptr[2]); + r = MAXJSAMPLE - RANGE_LIMIT(inptr[0]); + g = MAXJSAMPLE - RANGE_LIMIT(inptr[1]); + b = MAXJSAMPLE - RANGE_LIMIT(inptr[2]); /* K passes through as-is */ - outptr3[col] = inptr[3]; /* don't need GETJSAMPLE here */ + outptr3[col] = inptr[3]; inptr += 4; /* If the inputs are 0..MAXJSAMPLE, the outputs of these equations * must be too; we do not need an explicit range-limiting operation. @@ -438,7 +449,7 @@ grayscale_convert(j_compress_ptr cinfo, JSAMPARRAY input_buf, outptr = output_buf[0][output_row]; output_row++; for (col = 0; col < num_cols; col++) { - outptr[col] = inptr[0]; /* don't need GETJSAMPLE() here */ + outptr[col] = inptr[0]; inptr += instride; } } @@ -497,7 +508,7 @@ null_convert(j_compress_ptr cinfo, JSAMPARRAY input_buf, JSAMPIMAGE output_buf, inptr = *input_buf; outptr = output_buf[ci][output_row]; for (col = 0; col < num_cols; col++) { - outptr[col] = inptr[ci]; /* don't need GETJSAMPLE() here */ + outptr[col] = inptr[ci]; inptr += nc; } } diff --git a/src/3rdparty/libjpeg/src/jcdctmgr.c b/src/3rdparty/libjpeg/src/jcdctmgr.c index c04058e6ce..7dae17a6e1 100644 --- a/src/3rdparty/libjpeg/src/jcdctmgr.c +++ b/src/3rdparty/libjpeg/src/jcdctmgr.c @@ -381,19 +381,19 @@ convsamp(JSAMPARRAY sample_data, JDIMENSION start_col, DCTELEM *workspace) elemptr = sample_data[elemr] + start_col; #if DCTSIZE == 8 /* unroll the inner loop */ - *workspaceptr++ = GETJSAMPLE(*elemptr++) - CENTERJSAMPLE; - *workspaceptr++ = GETJSAMPLE(*elemptr++) - CENTERJSAMPLE; - *workspaceptr++ = GETJSAMPLE(*elemptr++) - CENTERJSAMPLE; - *workspaceptr++ = GETJSAMPLE(*elemptr++) - CENTERJSAMPLE; - *workspaceptr++ = GETJSAMPLE(*elemptr++) - CENTERJSAMPLE; - *workspaceptr++ = GETJSAMPLE(*elemptr++) - CENTERJSAMPLE; - *workspaceptr++ = GETJSAMPLE(*elemptr++) - CENTERJSAMPLE; - *workspaceptr++ = GETJSAMPLE(*elemptr++) - CENTERJSAMPLE; + *workspaceptr++ = (*elemptr++) - CENTERJSAMPLE; + *workspaceptr++ = (*elemptr++) - CENTERJSAMPLE; + *workspaceptr++ = (*elemptr++) - CENTERJSAMPLE; + *workspaceptr++ = (*elemptr++) - CENTERJSAMPLE; + *workspaceptr++ = (*elemptr++) - CENTERJSAMPLE; + *workspaceptr++ = (*elemptr++) - CENTERJSAMPLE; + *workspaceptr++ = (*elemptr++) - CENTERJSAMPLE; + *workspaceptr++ = (*elemptr++) - CENTERJSAMPLE; #else { register int elemc; for (elemc = DCTSIZE; elemc > 0; elemc--) - *workspaceptr++ = GETJSAMPLE(*elemptr++) - CENTERJSAMPLE; + *workspaceptr++ = (*elemptr++) - CENTERJSAMPLE; } #endif } @@ -533,20 +533,19 @@ convsamp_float(JSAMPARRAY sample_data, JDIMENSION start_col, for (elemr = 0; elemr < DCTSIZE; elemr++) { elemptr = sample_data[elemr] + start_col; #if DCTSIZE == 8 /* unroll the inner loop */ - *workspaceptr++ = (FAST_FLOAT)(GETJSAMPLE(*elemptr++) - CENTERJSAMPLE); - *workspaceptr++ = (FAST_FLOAT)(GETJSAMPLE(*elemptr++) - CENTERJSAMPLE); - *workspaceptr++ = (FAST_FLOAT)(GETJSAMPLE(*elemptr++) - CENTERJSAMPLE); - *workspaceptr++ = (FAST_FLOAT)(GETJSAMPLE(*elemptr++) - CENTERJSAMPLE); - *workspaceptr++ = (FAST_FLOAT)(GETJSAMPLE(*elemptr++) - CENTERJSAMPLE); - *workspaceptr++ = (FAST_FLOAT)(GETJSAMPLE(*elemptr++) - CENTERJSAMPLE); - *workspaceptr++ = (FAST_FLOAT)(GETJSAMPLE(*elemptr++) - CENTERJSAMPLE); - *workspaceptr++ = (FAST_FLOAT)(GETJSAMPLE(*elemptr++) - CENTERJSAMPLE); + *workspaceptr++ = (FAST_FLOAT)((*elemptr++) - CENTERJSAMPLE); + *workspaceptr++ = (FAST_FLOAT)((*elemptr++) - CENTERJSAMPLE); + *workspaceptr++ = (FAST_FLOAT)((*elemptr++) - CENTERJSAMPLE); + *workspaceptr++ = (FAST_FLOAT)((*elemptr++) - CENTERJSAMPLE); + *workspaceptr++ = (FAST_FLOAT)((*elemptr++) - CENTERJSAMPLE); + *workspaceptr++ = (FAST_FLOAT)((*elemptr++) - CENTERJSAMPLE); + *workspaceptr++ = (FAST_FLOAT)((*elemptr++) - CENTERJSAMPLE); + *workspaceptr++ = (FAST_FLOAT)((*elemptr++) - CENTERJSAMPLE); #else { register int elemc; for (elemc = DCTSIZE; elemc > 0; elemc--) - *workspaceptr++ = (FAST_FLOAT) - (GETJSAMPLE(*elemptr++) - CENTERJSAMPLE); + *workspaceptr++ = (FAST_FLOAT)((*elemptr++) - CENTERJSAMPLE); } #endif } diff --git a/src/3rdparty/libjpeg/src/jchuff.c b/src/3rdparty/libjpeg/src/jchuff.c index cb05055d99..5d0276ad25 100644 --- a/src/3rdparty/libjpeg/src/jchuff.c +++ b/src/3rdparty/libjpeg/src/jchuff.c @@ -4,8 +4,10 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1997, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2009-2011, 2014-2016, 2018-2019, D. R. Commander. + * Copyright (C) 2009-2011, 2014-2016, 2018-2022, D. R. Commander. * Copyright (C) 2015, Matthieu Darbois. + * Copyright (C) 2018, Matthias Räncker. + * Copyright (C) 2020, Arm Limited. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -25,7 +27,6 @@ #include "jinclude.h" #include "jpeglib.h" #include "jsimd.h" -#include "jconfigint.h" #include <limits.h> /* @@ -34,23 +35,28 @@ * memory footprint by 64k, which is important for some mobile applications * that create many isolated instances of libjpeg-turbo (web browsers, for * instance.) This may improve performance on some mobile platforms as well. - * This feature is enabled by default only on ARM processors, because some x86 + * This feature is enabled by default only on Arm processors, because some x86 * chips have a slow implementation of bsr, and the use of clz/bsr cannot be * shown to have a significant performance impact even on the x86 chips that - * have a fast implementation of it. When building for ARMv6, you can + * have a fast implementation of it. When building for Armv6, you can * explicitly disable the use of clz/bsr by adding -mthumb to the compiler * flags (this defines __thumb__). */ /* NOTE: Both GCC and Clang define __GNUC__ */ -#if defined(__GNUC__) && (defined(__arm__) || defined(__aarch64__)) +#if (defined(__GNUC__) && (defined(__arm__) || defined(__aarch64__))) || \ + defined(_M_ARM) || defined(_M_ARM64) #if !defined(__thumb__) || defined(__thumb2__) #define USE_CLZ_INTRINSIC #endif #endif #ifdef USE_CLZ_INTRINSIC +#if defined(_MSC_VER) && !defined(__clang__) +#define JPEG_NBITS_NONZERO(x) (32 - _CountLeadingZeros(x)) +#else #define JPEG_NBITS_NONZERO(x) (32 - __builtin_clz(x)) +#endif #define JPEG_NBITS(x) (x ? JPEG_NBITS_NONZERO(x) : 0) #else #include "jpeg_nbits_table.h" @@ -65,31 +71,42 @@ * but must not be updated permanently until we complete the MCU. */ -typedef struct { - size_t put_buffer; /* current bit-accumulation buffer */ - int put_bits; /* # of bits now in it */ - int last_dc_val[MAX_COMPS_IN_SCAN]; /* last DC coef for each component */ -} savable_state; +#if defined(__x86_64__) && defined(__ILP32__) +typedef unsigned long long bit_buf_type; +#else +typedef size_t bit_buf_type; +#endif -/* This macro is to work around compilers with missing or broken - * structure assignment. You'll need to fix this code if you have - * such a compiler and you change MAX_COMPS_IN_SCAN. +/* NOTE: The more optimal Huffman encoding algorithm is only used by the + * intrinsics implementation of the Arm Neon SIMD extensions, which is why we + * retain the old Huffman encoder behavior when using the GAS implementation. */ - -#ifndef NO_STRUCT_ASSIGN -#define ASSIGN_STATE(dest, src) ((dest) = (src)) +#if defined(WITH_SIMD) && !(defined(__arm__) || defined(__aarch64__) || \ + defined(_M_ARM) || defined(_M_ARM64)) +typedef unsigned long long simd_bit_buf_type; #else -#if MAX_COMPS_IN_SCAN == 4 -#define ASSIGN_STATE(dest, src) \ - ((dest).put_buffer = (src).put_buffer, \ - (dest).put_bits = (src).put_bits, \ - (dest).last_dc_val[0] = (src).last_dc_val[0], \ - (dest).last_dc_val[1] = (src).last_dc_val[1], \ - (dest).last_dc_val[2] = (src).last_dc_val[2], \ - (dest).last_dc_val[3] = (src).last_dc_val[3]) +typedef bit_buf_type simd_bit_buf_type; #endif + +#if (defined(SIZEOF_SIZE_T) && SIZEOF_SIZE_T == 8) || defined(_WIN64) || \ + (defined(__x86_64__) && defined(__ILP32__)) +#define BIT_BUF_SIZE 64 +#elif (defined(SIZEOF_SIZE_T) && SIZEOF_SIZE_T == 4) || defined(_WIN32) +#define BIT_BUF_SIZE 32 +#else +#error Cannot determine word size #endif +#define SIMD_BIT_BUF_SIZE (sizeof(simd_bit_buf_type) * 8) +typedef struct { + union { + bit_buf_type c; + simd_bit_buf_type simd; + } put_buffer; /* current bit accumulation buffer */ + int free_bits; /* # of bits available in it */ + /* (Neon GAS: # of bits now in it) */ + int last_dc_val[MAX_COMPS_IN_SCAN]; /* last DC coef for each component */ +} savable_state; typedef struct { struct jpeg_entropy_encoder pub; /* public fields */ @@ -123,6 +140,7 @@ typedef struct { size_t free_in_buffer; /* # of byte spaces remaining in buffer */ savable_state cur; /* Current bit buffer & DC state */ j_compress_ptr cinfo; /* dump_buffer needs access to this */ + int simd; } working_state; @@ -181,12 +199,12 @@ start_pass_huff(j_compress_ptr cinfo, boolean gather_statistics) entropy->dc_count_ptrs[dctbl] = (long *) (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, 257 * sizeof(long)); - MEMZERO(entropy->dc_count_ptrs[dctbl], 257 * sizeof(long)); + memset(entropy->dc_count_ptrs[dctbl], 0, 257 * sizeof(long)); if (entropy->ac_count_ptrs[actbl] == NULL) entropy->ac_count_ptrs[actbl] = (long *) (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, 257 * sizeof(long)); - MEMZERO(entropy->ac_count_ptrs[actbl], 257 * sizeof(long)); + memset(entropy->ac_count_ptrs[actbl], 0, 257 * sizeof(long)); #endif } else { /* Compute derived values for Huffman tables */ @@ -201,8 +219,17 @@ start_pass_huff(j_compress_ptr cinfo, boolean gather_statistics) } /* Initialize bit buffer to empty */ - entropy->saved.put_buffer = 0; - entropy->saved.put_bits = 0; + if (entropy->simd) { + entropy->saved.put_buffer.simd = 0; +#if defined(__aarch64__) && !defined(NEON_INTRINSICS) + entropy->saved.free_bits = 0; +#else + entropy->saved.free_bits = SIMD_BIT_BUF_SIZE; +#endif + } else { + entropy->saved.put_buffer.c = 0; + entropy->saved.free_bits = BIT_BUF_SIZE; + } /* Initialize restart stuff */ entropy->restarts_to_go = cinfo->restart_interval; @@ -287,7 +314,8 @@ jpeg_make_c_derived_tbl(j_compress_ptr cinfo, boolean isDC, int tblno, * this lets us detect duplicate VAL entries here, and later * allows emit_bits to detect any attempt to emit such symbols. */ - MEMZERO(dtbl->ehufsi, sizeof(dtbl->ehufsi)); + memset(dtbl->ehufco, 0, sizeof(dtbl->ehufco)); + memset(dtbl->ehufsi, 0, sizeof(dtbl->ehufsi)); /* This is also a convenient place to check for out-of-range * and duplicated VAL entries. We allow 0..255 for AC symbols @@ -334,94 +362,94 @@ dump_buffer(working_state *state) /* Outputting bits to the file */ -/* These macros perform the same task as the emit_bits() function in the - * original libjpeg code. In addition to reducing overhead by explicitly - * inlining the code, additional performance is achieved by taking into - * account the size of the bit buffer and waiting until it is almost full - * before emptying it. This mostly benefits 64-bit platforms, since 6 - * bytes can be stored in a 64-bit bit buffer before it has to be emptied. +/* Output byte b and, speculatively, an additional 0 byte. 0xFF must be + * encoded as 0xFF 0x00, so the output buffer pointer is advanced by 2 if the + * byte is 0xFF. Otherwise, the output buffer pointer is advanced by 1, and + * the speculative 0 byte will be overwritten by the next byte. */ - -#define EMIT_BYTE() { \ - JOCTET c; \ - put_bits -= 8; \ - c = (JOCTET)GETJOCTET(put_buffer >> put_bits); \ - *buffer++ = c; \ - if (c == 0xFF) /* need to stuff a zero byte? */ \ - *buffer++ = 0; \ +#define EMIT_BYTE(b) { \ + buffer[0] = (JOCTET)(b); \ + buffer[1] = 0; \ + buffer -= -2 + ((JOCTET)(b) < 0xFF); \ } -#define PUT_BITS(code, size) { \ - put_bits += size; \ - put_buffer = (put_buffer << size) | code; \ -} - -#if SIZEOF_SIZE_T != 8 && !defined(_WIN64) - -#define CHECKBUF15() { \ - if (put_bits > 15) { \ - EMIT_BYTE() \ - EMIT_BYTE() \ +/* Output the entire bit buffer. If there are no 0xFF bytes in it, then write + * directly to the output buffer. Otherwise, use the EMIT_BYTE() macro to + * encode 0xFF as 0xFF 0x00. + */ +#if BIT_BUF_SIZE == 64 + +#define FLUSH() { \ + if (put_buffer & 0x8080808080808080 & ~(put_buffer + 0x0101010101010101)) { \ + EMIT_BYTE(put_buffer >> 56) \ + EMIT_BYTE(put_buffer >> 48) \ + EMIT_BYTE(put_buffer >> 40) \ + EMIT_BYTE(put_buffer >> 32) \ + EMIT_BYTE(put_buffer >> 24) \ + EMIT_BYTE(put_buffer >> 16) \ + EMIT_BYTE(put_buffer >> 8) \ + EMIT_BYTE(put_buffer ) \ + } else { \ + buffer[0] = (JOCTET)(put_buffer >> 56); \ + buffer[1] = (JOCTET)(put_buffer >> 48); \ + buffer[2] = (JOCTET)(put_buffer >> 40); \ + buffer[3] = (JOCTET)(put_buffer >> 32); \ + buffer[4] = (JOCTET)(put_buffer >> 24); \ + buffer[5] = (JOCTET)(put_buffer >> 16); \ + buffer[6] = (JOCTET)(put_buffer >> 8); \ + buffer[7] = (JOCTET)(put_buffer); \ + buffer += 8; \ } \ } -#endif - -#define CHECKBUF31() { \ - if (put_bits > 31) { \ - EMIT_BYTE() \ - EMIT_BYTE() \ - EMIT_BYTE() \ - EMIT_BYTE() \ - } \ -} +#else -#define CHECKBUF47() { \ - if (put_bits > 47) { \ - EMIT_BYTE() \ - EMIT_BYTE() \ - EMIT_BYTE() \ - EMIT_BYTE() \ - EMIT_BYTE() \ - EMIT_BYTE() \ +#define FLUSH() { \ + if (put_buffer & 0x80808080 & ~(put_buffer + 0x01010101)) { \ + EMIT_BYTE(put_buffer >> 24) \ + EMIT_BYTE(put_buffer >> 16) \ + EMIT_BYTE(put_buffer >> 8) \ + EMIT_BYTE(put_buffer ) \ + } else { \ + buffer[0] = (JOCTET)(put_buffer >> 24); \ + buffer[1] = (JOCTET)(put_buffer >> 16); \ + buffer[2] = (JOCTET)(put_buffer >> 8); \ + buffer[3] = (JOCTET)(put_buffer); \ + buffer += 4; \ } \ } -#if !defined(_WIN32) && !defined(SIZEOF_SIZE_T) -#error Cannot determine word size #endif -#if SIZEOF_SIZE_T == 8 || defined(_WIN64) - -#define EMIT_BITS(code, size) { \ - CHECKBUF47() \ - PUT_BITS(code, size) \ -} - -#define EMIT_CODE(code, size) { \ - temp2 &= (((JLONG)1) << nbits) - 1; \ - CHECKBUF31() \ - PUT_BITS(code, size) \ - PUT_BITS(temp2, nbits) \ +/* Fill the bit buffer to capacity with the leading bits from code, then output + * the bit buffer and put the remaining bits from code into the bit buffer. + */ +#define PUT_AND_FLUSH(code, size) { \ + put_buffer = (put_buffer << (size + free_bits)) | (code >> -free_bits); \ + FLUSH() \ + free_bits += BIT_BUF_SIZE; \ + put_buffer = code; \ } -#else - -#define EMIT_BITS(code, size) { \ - PUT_BITS(code, size) \ - CHECKBUF15() \ +/* Insert code into the bit buffer and output the bit buffer if needed. + * NOTE: We can't flush with free_bits == 0, since the left shift in + * PUT_AND_FLUSH() would have undefined behavior. + */ +#define PUT_BITS(code, size) { \ + free_bits -= size; \ + if (free_bits < 0) \ + PUT_AND_FLUSH(code, size) \ + else \ + put_buffer = (put_buffer << size) | code; \ } -#define EMIT_CODE(code, size) { \ - temp2 &= (((JLONG)1) << nbits) - 1; \ - PUT_BITS(code, size) \ - CHECKBUF15() \ - PUT_BITS(temp2, nbits) \ - CHECKBUF15() \ +#define PUT_CODE(code, size) { \ + temp &= (((JLONG)1) << nbits) - 1; \ + temp |= code << nbits; \ + nbits += size; \ + PUT_BITS(temp, nbits) \ } -#endif - /* Although it is exceedingly rare, it is possible for a Huffman-encoded * coefficient block to be larger than the 128-byte unencoded block. For each @@ -444,11 +472,12 @@ dump_buffer(working_state *state) #define STORE_BUFFER() { \ if (localbuf) { \ + size_t bytes, bytestocopy; \ bytes = buffer - _buffer; \ buffer = _buffer; \ while (bytes > 0) { \ bytestocopy = MIN(bytes, state->free_in_buffer); \ - MEMCOPY(state->next_output_byte, buffer, bytestocopy); \ + memcpy(state->next_output_byte, buffer, bytestocopy); \ state->next_output_byte += bytestocopy; \ buffer += bytestocopy; \ state->free_in_buffer -= bytestocopy; \ @@ -466,20 +495,46 @@ dump_buffer(working_state *state) LOCAL(boolean) flush_bits(working_state *state) { - JOCTET _buffer[BUFSIZE], *buffer; - size_t put_buffer; int put_bits; - size_t bytes, bytestocopy; int localbuf = 0; + JOCTET _buffer[BUFSIZE], *buffer, temp; + simd_bit_buf_type put_buffer; int put_bits; + int localbuf = 0; + + if (state->simd) { +#if defined(__aarch64__) && !defined(NEON_INTRINSICS) + put_bits = state->cur.free_bits; +#else + put_bits = SIMD_BIT_BUF_SIZE - state->cur.free_bits; +#endif + put_buffer = state->cur.put_buffer.simd; + } else { + put_bits = BIT_BUF_SIZE - state->cur.free_bits; + put_buffer = state->cur.put_buffer.c; + } - put_buffer = state->cur.put_buffer; - put_bits = state->cur.put_bits; LOAD_BUFFER() - /* fill any partial byte with ones */ - PUT_BITS(0x7F, 7) - while (put_bits >= 8) EMIT_BYTE() + while (put_bits >= 8) { + put_bits -= 8; + temp = (JOCTET)(put_buffer >> put_bits); + EMIT_BYTE(temp) + } + if (put_bits) { + /* fill partial byte with ones */ + temp = (JOCTET)((put_buffer << (8 - put_bits)) | (0xFF >> put_bits)); + EMIT_BYTE(temp) + } - state->cur.put_buffer = 0; /* and reset bit-buffer to empty */ - state->cur.put_bits = 0; + if (state->simd) { /* and reset bit buffer to empty */ + state->cur.put_buffer.simd = 0; +#if defined(__aarch64__) && !defined(NEON_INTRINSICS) + state->cur.free_bits = 0; +#else + state->cur.free_bits = SIMD_BIT_BUF_SIZE; +#endif + } else { + state->cur.put_buffer.c = 0; + state->cur.free_bits = BIT_BUF_SIZE; + } STORE_BUFFER() return TRUE; @@ -493,7 +548,7 @@ encode_one_block_simd(working_state *state, JCOEFPTR block, int last_dc_val, c_derived_tbl *dctbl, c_derived_tbl *actbl) { JOCTET _buffer[BUFSIZE], *buffer; - size_t bytes, bytestocopy; int localbuf = 0; + int localbuf = 0; LOAD_BUFFER() @@ -509,53 +564,41 @@ LOCAL(boolean) encode_one_block(working_state *state, JCOEFPTR block, int last_dc_val, c_derived_tbl *dctbl, c_derived_tbl *actbl) { - int temp, temp2, temp3; - int nbits; - int r, code, size; + int temp, nbits, free_bits; + bit_buf_type put_buffer; JOCTET _buffer[BUFSIZE], *buffer; - size_t put_buffer; int put_bits; - int code_0xf0 = actbl->ehufco[0xf0], size_0xf0 = actbl->ehufsi[0xf0]; - size_t bytes, bytestocopy; int localbuf = 0; + int localbuf = 0; - put_buffer = state->cur.put_buffer; - put_bits = state->cur.put_bits; + free_bits = state->cur.free_bits; + put_buffer = state->cur.put_buffer.c; LOAD_BUFFER() /* Encode the DC coefficient difference per section F.1.2.1 */ - temp = temp2 = block[0] - last_dc_val; + temp = block[0] - last_dc_val; /* This is a well-known technique for obtaining the absolute value without a * branch. It is derived from an assembly language technique presented in * "How to Optimize for the Pentium Processors", Copyright (c) 1996, 1997 by - * Agner Fog. + * Agner Fog. This code assumes we are on a two's complement machine. */ - temp3 = temp >> (CHAR_BIT * sizeof(int) - 1); - temp ^= temp3; - temp -= temp3; - - /* For a negative input, want temp2 = bitwise complement of abs(input) */ - /* This code assumes we are on a two's complement machine */ - temp2 += temp3; + nbits = temp >> (CHAR_BIT * sizeof(int) - 1); + temp += nbits; + nbits ^= temp; /* Find the number of bits needed for the magnitude of the coefficient */ - nbits = JPEG_NBITS(temp); - - /* Emit the Huffman-coded symbol for the number of bits */ - code = dctbl->ehufco[nbits]; - size = dctbl->ehufsi[nbits]; - EMIT_BITS(code, size) + nbits = JPEG_NBITS(nbits); - /* Mask off any extra bits in code */ - temp2 &= (((JLONG)1) << nbits) - 1; - - /* Emit that number of bits of the value, if positive, */ - /* or the complement of its magnitude, if negative. */ - EMIT_BITS(temp2, nbits) + /* Emit the Huffman-coded symbol for the number of bits. + * Emit that number of bits of the value, if positive, + * or the complement of its magnitude, if negative. + */ + PUT_CODE(dctbl->ehufco[nbits], dctbl->ehufsi[nbits]) /* Encode the AC coefficients per section F.1.2.2 */ - r = 0; /* r = run length of zeros */ + { + int r = 0; /* r = run length of zeros */ /* Manually unroll the k loop to eliminate the counter variable. This * improves performance greatly on systems with a limited number of @@ -563,51 +606,46 @@ encode_one_block(working_state *state, JCOEFPTR block, int last_dc_val, */ #define kloop(jpeg_natural_order_of_k) { \ if ((temp = block[jpeg_natural_order_of_k]) == 0) { \ - r++; \ + r += 16; \ } else { \ - temp2 = temp; \ /* Branch-less absolute value, bitwise complement, etc., same as above */ \ - temp3 = temp >> (CHAR_BIT * sizeof(int) - 1); \ - temp ^= temp3; \ - temp -= temp3; \ - temp2 += temp3; \ - nbits = JPEG_NBITS_NONZERO(temp); \ + nbits = temp >> (CHAR_BIT * sizeof(int) - 1); \ + temp += nbits; \ + nbits ^= temp; \ + nbits = JPEG_NBITS_NONZERO(nbits); \ /* if run length > 15, must emit special run-length-16 codes (0xF0) */ \ - while (r > 15) { \ - EMIT_BITS(code_0xf0, size_0xf0) \ - r -= 16; \ + while (r >= 16 * 16) { \ + r -= 16 * 16; \ + PUT_BITS(actbl->ehufco[0xf0], actbl->ehufsi[0xf0]) \ } \ /* Emit Huffman symbol for run length / number of bits */ \ - temp3 = (r << 4) + nbits; \ - code = actbl->ehufco[temp3]; \ - size = actbl->ehufsi[temp3]; \ - EMIT_CODE(code, size) \ + r += nbits; \ + PUT_CODE(actbl->ehufco[r], actbl->ehufsi[r]) \ r = 0; \ } \ } - /* One iteration for each value in jpeg_natural_order[] */ - kloop(1); kloop(8); kloop(16); kloop(9); kloop(2); kloop(3); - kloop(10); kloop(17); kloop(24); kloop(32); kloop(25); kloop(18); - kloop(11); kloop(4); kloop(5); kloop(12); kloop(19); kloop(26); - kloop(33); kloop(40); kloop(48); kloop(41); kloop(34); kloop(27); - kloop(20); kloop(13); kloop(6); kloop(7); kloop(14); kloop(21); - kloop(28); kloop(35); kloop(42); kloop(49); kloop(56); kloop(57); - kloop(50); kloop(43); kloop(36); kloop(29); kloop(22); kloop(15); - kloop(23); kloop(30); kloop(37); kloop(44); kloop(51); kloop(58); - kloop(59); kloop(52); kloop(45); kloop(38); kloop(31); kloop(39); - kloop(46); kloop(53); kloop(60); kloop(61); kloop(54); kloop(47); - kloop(55); kloop(62); kloop(63); - - /* If the last coef(s) were zero, emit an end-of-block code */ - if (r > 0) { - code = actbl->ehufco[0]; - size = actbl->ehufsi[0]; - EMIT_BITS(code, size) + /* One iteration for each value in jpeg_natural_order[] */ + kloop(1); kloop(8); kloop(16); kloop(9); kloop(2); kloop(3); + kloop(10); kloop(17); kloop(24); kloop(32); kloop(25); kloop(18); + kloop(11); kloop(4); kloop(5); kloop(12); kloop(19); kloop(26); + kloop(33); kloop(40); kloop(48); kloop(41); kloop(34); kloop(27); + kloop(20); kloop(13); kloop(6); kloop(7); kloop(14); kloop(21); + kloop(28); kloop(35); kloop(42); kloop(49); kloop(56); kloop(57); + kloop(50); kloop(43); kloop(36); kloop(29); kloop(22); kloop(15); + kloop(23); kloop(30); kloop(37); kloop(44); kloop(51); kloop(58); + kloop(59); kloop(52); kloop(45); kloop(38); kloop(31); kloop(39); + kloop(46); kloop(53); kloop(60); kloop(61); kloop(54); kloop(47); + kloop(55); kloop(62); kloop(63); + + /* If the last coef(s) were zero, emit an end-of-block code */ + if (r > 0) { + PUT_BITS(actbl->ehufco[0], actbl->ehufsi[0]) + } } - state->cur.put_buffer = put_buffer; - state->cur.put_bits = put_bits; + state->cur.put_buffer.c = put_buffer; + state->cur.free_bits = free_bits; STORE_BUFFER() return TRUE; @@ -654,8 +692,9 @@ encode_mcu_huff(j_compress_ptr cinfo, JBLOCKROW *MCU_data) /* Load up working state */ state.next_output_byte = cinfo->dest->next_output_byte; state.free_in_buffer = cinfo->dest->free_in_buffer; - ASSIGN_STATE(state.cur, entropy->saved); + state.cur = entropy->saved; state.cinfo = cinfo; + state.simd = entropy->simd; /* Emit restart marker if needed */ if (cinfo->restart_interval) { @@ -694,7 +733,7 @@ encode_mcu_huff(j_compress_ptr cinfo, JBLOCKROW *MCU_data) /* Completed MCU, so update state */ cinfo->dest->next_output_byte = state.next_output_byte; cinfo->dest->free_in_buffer = state.free_in_buffer; - ASSIGN_STATE(entropy->saved, state.cur); + entropy->saved = state.cur; /* Update restart-interval state too */ if (cinfo->restart_interval) { @@ -723,8 +762,9 @@ finish_pass_huff(j_compress_ptr cinfo) /* Load up working state ... flush_bits needs it */ state.next_output_byte = cinfo->dest->next_output_byte; state.free_in_buffer = cinfo->dest->free_in_buffer; - ASSIGN_STATE(state.cur, entropy->saved); + state.cur = entropy->saved; state.cinfo = cinfo; + state.simd = entropy->simd; /* Flush out the last data */ if (!flush_bits(&state)) @@ -733,7 +773,7 @@ finish_pass_huff(j_compress_ptr cinfo) /* Update state */ cinfo->dest->next_output_byte = state.next_output_byte; cinfo->dest->free_in_buffer = state.free_in_buffer; - ASSIGN_STATE(entropy->saved, state.cur); + entropy->saved = state.cur; } @@ -900,8 +940,8 @@ jpeg_gen_optimal_table(j_compress_ptr cinfo, JHUFF_TBL *htbl, long freq[]) /* This algorithm is explained in section K.2 of the JPEG standard */ - MEMZERO(bits, sizeof(bits)); - MEMZERO(codesize, sizeof(codesize)); + memset(bits, 0, sizeof(bits)); + memset(codesize, 0, sizeof(codesize)); for (i = 0; i < 257; i++) others[i] = -1; /* init links to empty */ @@ -1003,7 +1043,7 @@ jpeg_gen_optimal_table(j_compress_ptr cinfo, JHUFF_TBL *htbl, long freq[]) bits[i]--; /* Return final symbol counts (only for lengths 0..16) */ - MEMCOPY(htbl->bits, bits, sizeof(htbl->bits)); + memcpy(htbl->bits, bits, sizeof(htbl->bits)); /* Return a list of the symbols sorted by code length */ /* It's not real clear to me why we don't need to consider the codelength @@ -1042,8 +1082,8 @@ finish_pass_gather(j_compress_ptr cinfo) /* It's important not to apply jpeg_gen_optimal_table more than once * per table, because it clobbers the input frequency counts! */ - MEMZERO(did_dc, sizeof(did_dc)); - MEMZERO(did_ac, sizeof(did_ac)); + memset(did_dc, 0, sizeof(did_dc)); + memset(did_ac, 0, sizeof(did_ac)); for (ci = 0; ci < cinfo->comps_in_scan; ci++) { compptr = cinfo->cur_comp_info[ci]; diff --git a/src/3rdparty/libjpeg/src/jchuff.h b/src/3rdparty/libjpeg/src/jchuff.h index 314a2325c9..da7809a94b 100644 --- a/src/3rdparty/libjpeg/src/jchuff.h +++ b/src/3rdparty/libjpeg/src/jchuff.h @@ -3,8 +3,8 @@ * * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1997, Thomas G. Lane. - * It was modified by The libjpeg-turbo Project to include only code relevant - * to libjpeg-turbo. + * libjpeg-turbo Modifications: + * Copyright (C) 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -25,6 +25,14 @@ #define MAX_COEF_BITS 14 #endif +/* The progressive Huffman encoder uses an unsigned 16-bit data type to store + * absolute values of coefficients, because it is possible to inject a + * coefficient value of -32768 into the encoder by attempting to transform a + * malformed 12-bit JPEG image, and the absolute value of -32768 would overflow + * a signed 16-bit integer. + */ +typedef unsigned short UJCOEF; + /* Derived data constructed for each Huffman table */ typedef struct { diff --git a/src/3rdparty/libjpeg/src/jcinit.c b/src/3rdparty/libjpeg/src/jcinit.c index 78aa465786..157353a22e 100644 --- a/src/3rdparty/libjpeg/src/jcinit.c +++ b/src/3rdparty/libjpeg/src/jcinit.c @@ -1,8 +1,10 @@ /* * jcinit.c * + * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1997, Thomas G. Lane. - * This file is part of the Independent JPEG Group's software. + * libjpeg-turbo Modifications: + * Copyright (C) 2020, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -19,6 +21,7 @@ #define JPEG_INTERNALS #include "jinclude.h" #include "jpeglib.h" +#include "jpegcomp.h" /* diff --git a/src/3rdparty/libjpeg/src/jcmaster.c b/src/3rdparty/libjpeg/src/jcmaster.c index 998dc40a5c..b821710ac3 100644 --- a/src/3rdparty/libjpeg/src/jcmaster.c +++ b/src/3rdparty/libjpeg/src/jcmaster.c @@ -19,7 +19,6 @@ #include "jinclude.h" #include "jpeglib.h" #include "jpegcomp.h" -#include "jconfigint.h" /* Private state */ @@ -493,7 +492,7 @@ prepare_for_pass(j_compress_ptr cinfo) master->pass_type = output_pass; master->pass_number++; #endif - /*FALLTHROUGH*/ + FALLTHROUGH /*FALLTHROUGH*/ case output_pass: /* Do a data-output pass. */ /* We need not repeat per-scan setup if prior optimization pass did it. */ diff --git a/src/3rdparty/libjpeg/src/jconfig.h b/src/3rdparty/libjpeg/src/jconfig.h new file mode 100644 index 0000000000..51050d0434 --- /dev/null +++ b/src/3rdparty/libjpeg/src/jconfig.h @@ -0,0 +1,17 @@ +// Definitions for building in Qt source, ref. src/jconfig.h.in + +#define JPEG_LIB_VERSION 80 + +#define LIBJPEG_TURBO_VERSION 2.1.5 + +#define LIBJPEG_TURBO_VERSION_NUMBER 2001005 + +#define C_ARITH_CODING_SUPPORTED 1 + +#define D_ARITH_CODING_SUPPORTED 1 + +#define MEM_SRCDST_SUPPORTED 1 + +#define BITS_IN_JSAMPLE 8 + +#define NO_PUTENV diff --git a/src/3rdparty/libjpeg/src/jconfig.h.in b/src/3rdparty/libjpeg/src/jconfig.h.in index 18a69a4814..e0180122fe 100644 --- a/src/3rdparty/libjpeg/src/jconfig.h.in +++ b/src/3rdparty/libjpeg/src/jconfig.h.in @@ -32,42 +32,6 @@ #define BITS_IN_JSAMPLE @BITS_IN_JSAMPLE@ /* use 8 or 12 */ -/* Define to 1 if you have the <locale.h> header file. */ -#cmakedefine HAVE_LOCALE_H 1 - -/* Define to 1 if you have the <stddef.h> header file. */ -#cmakedefine HAVE_STDDEF_H 1 - -/* Define to 1 if you have the <stdlib.h> header file. */ -#cmakedefine HAVE_STDLIB_H 1 - -/* Define if you need to include <sys/types.h> to get size_t. */ -#cmakedefine NEED_SYS_TYPES_H 1 - -/* Define if you have BSD-like bzero and bcopy in <strings.h> rather than - memset/memcpy in <string.h>. */ -#cmakedefine NEED_BSD_STRINGS 1 - -/* Define to 1 if the system has the type `unsigned char'. */ -#cmakedefine HAVE_UNSIGNED_CHAR 1 - -/* Define to 1 if the system has the type `unsigned short'. */ -#cmakedefine HAVE_UNSIGNED_SHORT 1 - -/* Compiler does not support pointers to undefined structures. */ -#cmakedefine INCOMPLETE_TYPES_BROKEN 1 - /* Define if your (broken) compiler shifts signed values as if they were unsigned. */ #cmakedefine RIGHT_SHIFT_IS_UNSIGNED 1 - -/* Define to 1 if type `char' is unsigned and you are not using gcc. */ -#ifndef __CHAR_UNSIGNED__ - #cmakedefine __CHAR_UNSIGNED__ 1 -#endif - -/* Define to empty if `const' does not conform to ANSI C. */ -/* #undef const */ - -/* Define to `unsigned int' if <sys/types.h> does not define. */ -/* #undef size_t */ diff --git a/src/3rdparty/libjpeg/src/jconfigint.h b/src/3rdparty/libjpeg/src/jconfigint.h new file mode 100644 index 0000000000..97edb31963 --- /dev/null +++ b/src/3rdparty/libjpeg/src/jconfigint.h @@ -0,0 +1,19 @@ +// Definitions for building in Qt source, ref. src/jconfigint.h.in + +#include <stdint.h> + +#define BUILD "" + +#define INLINE inline + +#define PACKAGE_NAME "libjpeg-turbo" + +#define VERSION "2.1.5" + +#if SIZE_MAX == 0xffffffff +#define SIZEOF_SIZE_T 4 +#elif SIZE_MAX == 0xffffffffffffffff +#define SIZEOF_SIZE_T 8 +#endif + +#define FALLTHROUGH diff --git a/src/3rdparty/libjpeg/src/jconfigint.h.in b/src/3rdparty/libjpeg/src/jconfigint.h.in index 68cbc2a505..d087d7b553 100644 --- a/src/3rdparty/libjpeg/src/jconfigint.h.in +++ b/src/3rdparty/libjpeg/src/jconfigint.h.in @@ -32,3 +32,13 @@ #define HAVE_BITSCANFORWARD #endif #endif + +#if defined(__has_attribute) +#if __has_attribute(fallthrough) +#define FALLTHROUGH __attribute__((fallthrough)); +#else +#define FALLTHROUGH +#endif +#else +#define FALLTHROUGH +#endif diff --git a/src/3rdparty/libjpeg/src/jcphuff.c b/src/3rdparty/libjpeg/src/jcphuff.c index 8c4efaf16c..5006b67075 100644 --- a/src/3rdparty/libjpeg/src/jcphuff.c +++ b/src/3rdparty/libjpeg/src/jcphuff.c @@ -4,8 +4,10 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1995-1997, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2011, 2015, 2018, D. R. Commander. - * Copyright (C) 2016, 2018, Matthieu Darbois. + * Copyright (C) 2011, 2015, 2018, 2021-2022, D. R. Commander. + * Copyright (C) 2016, 2018, 2022, Matthieu Darbois. + * Copyright (C) 2020, Arm Limited. + * Copyright (C) 2021, Alex Richardson. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -20,7 +22,6 @@ #include "jinclude.h" #include "jpeglib.h" #include "jsimd.h" -#include "jconfigint.h" #include <limits.h> #ifdef HAVE_INTRIN_H @@ -43,23 +44,28 @@ * memory footprint by 64k, which is important for some mobile applications * that create many isolated instances of libjpeg-turbo (web browsers, for * instance.) This may improve performance on some mobile platforms as well. - * This feature is enabled by default only on ARM processors, because some x86 + * This feature is enabled by default only on Arm processors, because some x86 * chips have a slow implementation of bsr, and the use of clz/bsr cannot be * shown to have a significant performance impact even on the x86 chips that - * have a fast implementation of it. When building for ARMv6, you can + * have a fast implementation of it. When building for Armv6, you can * explicitly disable the use of clz/bsr by adding -mthumb to the compiler * flags (this defines __thumb__). */ /* NOTE: Both GCC and Clang define __GNUC__ */ -#if defined(__GNUC__) && (defined(__arm__) || defined(__aarch64__)) +#if (defined(__GNUC__) && (defined(__arm__) || defined(__aarch64__))) || \ + defined(_M_ARM) || defined(_M_ARM64) #if !defined(__thumb__) || defined(__thumb2__) #define USE_CLZ_INTRINSIC #endif #endif #ifdef USE_CLZ_INTRINSIC +#if defined(_MSC_VER) && !defined(__clang__) +#define JPEG_NBITS_NONZERO(x) (32 - _CountLeadingZeros(x)) +#else #define JPEG_NBITS_NONZERO(x) (32 - __builtin_clz(x)) +#endif #define JPEG_NBITS(x) (x ? JPEG_NBITS_NONZERO(x) : 0) #else #include "jpeg_nbits_table.h" @@ -76,11 +82,11 @@ typedef struct { /* Pointer to routine to prepare data for encode_mcu_AC_first() */ void (*AC_first_prepare) (const JCOEF *block, const int *jpeg_natural_order_start, int Sl, - int Al, JCOEF *values, size_t *zerobits); + int Al, UJCOEF *values, size_t *zerobits); /* Pointer to routine to prepare data for encode_mcu_AC_refine() */ int (*AC_refine_prepare) (const JCOEF *block, const int *jpeg_natural_order_start, int Sl, - int Al, JCOEF *absvalues, size_t *bits); + int Al, UJCOEF *absvalues, size_t *bits); /* Mode flag: TRUE for optimization, FALSE for actual data output */ boolean gather_statistics; @@ -150,14 +156,14 @@ METHODDEF(boolean) encode_mcu_DC_first(j_compress_ptr cinfo, JBLOCKROW *MCU_data); METHODDEF(void) encode_mcu_AC_first_prepare (const JCOEF *block, const int *jpeg_natural_order_start, int Sl, int Al, - JCOEF *values, size_t *zerobits); + UJCOEF *values, size_t *zerobits); METHODDEF(boolean) encode_mcu_AC_first(j_compress_ptr cinfo, JBLOCKROW *MCU_data); METHODDEF(boolean) encode_mcu_DC_refine(j_compress_ptr cinfo, JBLOCKROW *MCU_data); METHODDEF(int) encode_mcu_AC_refine_prepare (const JCOEF *block, const int *jpeg_natural_order_start, int Sl, int Al, - JCOEF *absvalues, size_t *bits); + UJCOEF *absvalues, size_t *bits); METHODDEF(boolean) encode_mcu_AC_refine(j_compress_ptr cinfo, JBLOCKROW *MCU_data); METHODDEF(void) finish_pass_phuff(j_compress_ptr cinfo); @@ -169,24 +175,26 @@ INLINE METHODDEF(int) count_zeroes(size_t *x) { - int result; #if defined(HAVE_BUILTIN_CTZL) + int result; result = __builtin_ctzl(*x); *x >>= result; #elif defined(HAVE_BITSCANFORWARD64) + unsigned long result; _BitScanForward64(&result, *x); *x >>= result; #elif defined(HAVE_BITSCANFORWARD) + unsigned long result; _BitScanForward(&result, *x); *x >>= result; #else - result = 0; + int result = 0; while ((*x & 1) == 0) { ++result; *x >>= 1; } #endif - return result; + return (int)result; } @@ -266,7 +274,7 @@ start_pass_phuff(j_compress_ptr cinfo, boolean gather_statistics) entropy->count_ptrs[tbl] = (long *) (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, 257 * sizeof(long)); - MEMZERO(entropy->count_ptrs[tbl], 257 * sizeof(long)); + memset(entropy->count_ptrs[tbl], 0, 257 * sizeof(long)); } else { /* Compute derived values for Huffman table */ /* We may do this more than once for a table, but it's not expensive */ @@ -575,8 +583,8 @@ encode_mcu_DC_first(j_compress_ptr cinfo, JBLOCKROW *MCU_data) continue; \ /* For a negative coef, want temp2 = bitwise complement of abs(coef) */ \ temp2 ^= temp; \ - values[k] = temp; \ - values[k + DCTSIZE2] = temp2; \ + values[k] = (UJCOEF)temp; \ + values[k + DCTSIZE2] = (UJCOEF)temp2; \ zerobits |= ((size_t)1U) << k; \ } \ } @@ -584,7 +592,7 @@ encode_mcu_DC_first(j_compress_ptr cinfo, JBLOCKROW *MCU_data) METHODDEF(void) encode_mcu_AC_first_prepare(const JCOEF *block, const int *jpeg_natural_order_start, int Sl, - int Al, JCOEF *values, size_t *bits) + int Al, UJCOEF *values, size_t *bits) { register int k, temp, temp2; size_t zerobits = 0U; @@ -657,9 +665,9 @@ encode_mcu_AC_first(j_compress_ptr cinfo, JBLOCKROW *MCU_data) register int nbits, r; int Sl = cinfo->Se - cinfo->Ss + 1; int Al = cinfo->Al; - JCOEF values_unaligned[2 * DCTSIZE2 + 15]; - JCOEF *values; - const JCOEF *cvalue; + UJCOEF values_unaligned[2 * DCTSIZE2 + 15]; + UJCOEF *values; + const UJCOEF *cvalue; size_t zerobits; size_t bits[8 / SIZEOF_SIZE_T]; @@ -672,7 +680,7 @@ encode_mcu_AC_first(j_compress_ptr cinfo, JBLOCKROW *MCU_data) emit_restart(entropy, entropy->next_restart_num); #ifdef WITH_SIMD - cvalue = values = (JCOEF *)PAD((size_t)values_unaligned, 16); + cvalue = values = (UJCOEF *)PAD((JUINTPTR)values_unaligned, 16); #else /* Not using SIMD, so alignment is not needed */ cvalue = values = values_unaligned; @@ -806,7 +814,7 @@ encode_mcu_DC_refine(j_compress_ptr cinfo, JBLOCKROW *MCU_data) zerobits |= ((size_t)1U) << k; \ signbits |= ((size_t)(temp2 + 1)) << k; \ } \ - absvalues[k] = (JCOEF)temp; /* save abs value for main pass */ \ + absvalues[k] = (UJCOEF)temp; /* save abs value for main pass */ \ if (temp == 1) \ EOB = k + koffset; /* EOB = index of last newly-nonzero coef */ \ } \ @@ -815,7 +823,7 @@ encode_mcu_DC_refine(j_compress_ptr cinfo, JBLOCKROW *MCU_data) METHODDEF(int) encode_mcu_AC_refine_prepare(const JCOEF *block, const int *jpeg_natural_order_start, int Sl, - int Al, JCOEF *absvalues, size_t *bits) + int Al, UJCOEF *absvalues, size_t *bits) { register int k, temp, temp2; int EOB = 0; @@ -860,7 +868,7 @@ encode_mcu_AC_refine_prepare(const JCOEF *block, #define ENCODE_COEFS_AC_REFINE(label) { \ while (zerobits) { \ - int idx = count_zeroes(&zerobits); \ + idx = count_zeroes(&zerobits); \ r += idx; \ cabsvalue += idx; \ signbits >>= idx; \ @@ -917,14 +925,14 @@ METHODDEF(boolean) encode_mcu_AC_refine(j_compress_ptr cinfo, JBLOCKROW *MCU_data) { phuff_entropy_ptr entropy = (phuff_entropy_ptr)cinfo->entropy; - register int temp, r; + register int temp, r, idx; char *BR_buffer; unsigned int BR; int Sl = cinfo->Se - cinfo->Ss + 1; int Al = cinfo->Al; - JCOEF absvalues_unaligned[DCTSIZE2 + 15]; - JCOEF *absvalues; - const JCOEF *cabsvalue, *EOBPTR; + UJCOEF absvalues_unaligned[DCTSIZE2 + 15]; + UJCOEF *absvalues; + const UJCOEF *cabsvalue, *EOBPTR; size_t zerobits, signbits; size_t bits[16 / SIZEOF_SIZE_T]; @@ -937,7 +945,7 @@ encode_mcu_AC_refine(j_compress_ptr cinfo, JBLOCKROW *MCU_data) emit_restart(entropy, entropy->next_restart_num); #ifdef WITH_SIMD - cabsvalue = absvalues = (JCOEF *)PAD((size_t)absvalues_unaligned, 16); + cabsvalue = absvalues = (UJCOEF *)PAD((JUINTPTR)absvalues_unaligned, 16); #else /* Not using SIMD, so alignment is not needed */ cabsvalue = absvalues = absvalues_unaligned; @@ -968,7 +976,7 @@ encode_mcu_AC_refine(j_compress_ptr cinfo, JBLOCKROW *MCU_data) if (zerobits) { int diff = ((absvalues + DCTSIZE2 / 2) - cabsvalue); - int idx = count_zeroes(&zerobits); + idx = count_zeroes(&zerobits); signbits >>= idx; idx += diff; r += idx; @@ -1053,7 +1061,7 @@ finish_pass_gather_phuff(j_compress_ptr cinfo) /* It's important not to apply jpeg_gen_optimal_table more than once * per table, because it clobbers the input frequency counts! */ - MEMZERO(did, sizeof(did)); + memset(did, 0, sizeof(did)); for (ci = 0; ci < cinfo->comps_in_scan; ci++) { compptr = cinfo->cur_comp_info[ci]; diff --git a/src/3rdparty/libjpeg/src/jcprepct.c b/src/3rdparty/libjpeg/src/jcprepct.c index d59713ae68..f27cc34507 100644 --- a/src/3rdparty/libjpeg/src/jcprepct.c +++ b/src/3rdparty/libjpeg/src/jcprepct.c @@ -3,8 +3,8 @@ * * This file is part of the Independent JPEG Group's software: * Copyright (C) 1994-1996, Thomas G. Lane. - * It was modified by The libjpeg-turbo Project to include only code relevant - * to libjpeg-turbo. + * libjpeg-turbo Modifications: + * Copyright (C) 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -289,8 +289,8 @@ create_context_buffer(j_compress_ptr cinfo) cinfo->max_h_samp_factor) / compptr->h_samp_factor), (JDIMENSION)(3 * rgroup_height)); /* Copy true buffer row pointers into the middle of the fake row array */ - MEMCOPY(fake_buffer + rgroup_height, true_buffer, - 3 * rgroup_height * sizeof(JSAMPROW)); + memcpy(fake_buffer + rgroup_height, true_buffer, + 3 * rgroup_height * sizeof(JSAMPROW)); /* Fill in the above and below wraparound pointers */ for (i = 0; i < rgroup_height; i++) { fake_buffer[i] = true_buffer[2 * rgroup_height + i]; diff --git a/src/3rdparty/libjpeg/src/jcsample.c b/src/3rdparty/libjpeg/src/jcsample.c index bd27b84e06..e8515ebf0f 100644 --- a/src/3rdparty/libjpeg/src/jcsample.c +++ b/src/3rdparty/libjpeg/src/jcsample.c @@ -6,7 +6,7 @@ * libjpeg-turbo Modifications: * Copyright 2009 Pierre Ossman <ossman@cendio.se> for Cendio AB * Copyright (C) 2014, MIPS Technologies, Inc., California. - * Copyright (C) 2015, D. R. Commander. + * Copyright (C) 2015, 2019, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -103,7 +103,7 @@ expand_right_edge(JSAMPARRAY image_data, int num_rows, JDIMENSION input_cols, if (numcols > 0) { for (row = 0; row < num_rows; row++) { ptr = image_data[row] + input_cols; - pixval = ptr[-1]; /* don't need GETJSAMPLE() here */ + pixval = ptr[-1]; for (count = numcols; count > 0; count--) *ptr++ = pixval; } @@ -174,7 +174,7 @@ int_downsample(j_compress_ptr cinfo, jpeg_component_info *compptr, for (v = 0; v < v_expand; v++) { inptr = input_data[inrow + v] + outcol_h; for (h = 0; h < h_expand; h++) { - outvalue += (JLONG)GETJSAMPLE(*inptr++); + outvalue += (JLONG)(*inptr++); } } *outptr++ = (JSAMPLE)((outvalue + numpix2) / numpix); @@ -237,8 +237,7 @@ h2v1_downsample(j_compress_ptr cinfo, jpeg_component_info *compptr, inptr = input_data[outrow]; bias = 0; /* bias = 0,1,0,1,... for successive samples */ for (outcol = 0; outcol < output_cols; outcol++) { - *outptr++ = - (JSAMPLE)((GETJSAMPLE(*inptr) + GETJSAMPLE(inptr[1]) + bias) >> 1); + *outptr++ = (JSAMPLE)((inptr[0] + inptr[1] + bias) >> 1); bias ^= 1; /* 0=>1, 1=>0 */ inptr += 2; } @@ -277,8 +276,7 @@ h2v2_downsample(j_compress_ptr cinfo, jpeg_component_info *compptr, bias = 1; /* bias = 1,2,1,2,... for successive samples */ for (outcol = 0; outcol < output_cols; outcol++) { *outptr++ = - (JSAMPLE)((GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]) + bias) >> 2); + (JSAMPLE)((inptr0[0] + inptr0[1] + inptr1[0] + inptr1[1] + bias) >> 2); bias ^= 3; /* 1=>2, 2=>1 */ inptr0 += 2; inptr1 += 2; } @@ -337,33 +335,25 @@ h2v2_smooth_downsample(j_compress_ptr cinfo, jpeg_component_info *compptr, below_ptr = input_data[inrow + 2]; /* Special case for first column: pretend column -1 is same as column 0 */ - membersum = GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]); - neighsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[1]) + - GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) + - GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[2]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[2]); + membersum = inptr0[0] + inptr0[1] + inptr1[0] + inptr1[1]; + neighsum = above_ptr[0] + above_ptr[1] + below_ptr[0] + below_ptr[1] + + inptr0[0] + inptr0[2] + inptr1[0] + inptr1[2]; neighsum += neighsum; - neighsum += GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[2]) + - GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[2]); + neighsum += above_ptr[0] + above_ptr[2] + below_ptr[0] + below_ptr[2]; membersum = membersum * memberscale + neighsum * neighscale; *outptr++ = (JSAMPLE)((membersum + 32768) >> 16); inptr0 += 2; inptr1 += 2; above_ptr += 2; below_ptr += 2; for (colctr = output_cols - 2; colctr > 0; colctr--) { /* sum of pixels directly mapped to this output element */ - membersum = GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]); + membersum = inptr0[0] + inptr0[1] + inptr1[0] + inptr1[1]; /* sum of edge-neighbor pixels */ - neighsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[1]) + - GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) + - GETJSAMPLE(inptr0[-1]) + GETJSAMPLE(inptr0[2]) + - GETJSAMPLE(inptr1[-1]) + GETJSAMPLE(inptr1[2]); + neighsum = above_ptr[0] + above_ptr[1] + below_ptr[0] + below_ptr[1] + + inptr0[-1] + inptr0[2] + inptr1[-1] + inptr1[2]; /* The edge-neighbors count twice as much as corner-neighbors */ neighsum += neighsum; /* Add in the corner-neighbors */ - neighsum += GETJSAMPLE(above_ptr[-1]) + GETJSAMPLE(above_ptr[2]) + - GETJSAMPLE(below_ptr[-1]) + GETJSAMPLE(below_ptr[2]); + neighsum += above_ptr[-1] + above_ptr[2] + below_ptr[-1] + below_ptr[2]; /* form final output scaled up by 2^16 */ membersum = membersum * memberscale + neighsum * neighscale; /* round, descale and output it */ @@ -372,15 +362,11 @@ h2v2_smooth_downsample(j_compress_ptr cinfo, jpeg_component_info *compptr, } /* Special case for last column */ - membersum = GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]); - neighsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[1]) + - GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) + - GETJSAMPLE(inptr0[-1]) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(inptr1[-1]) + GETJSAMPLE(inptr1[1]); + membersum = inptr0[0] + inptr0[1] + inptr1[0] + inptr1[1]; + neighsum = above_ptr[0] + above_ptr[1] + below_ptr[0] + below_ptr[1] + + inptr0[-1] + inptr0[1] + inptr1[-1] + inptr1[1]; neighsum += neighsum; - neighsum += GETJSAMPLE(above_ptr[-1]) + GETJSAMPLE(above_ptr[1]) + - GETJSAMPLE(below_ptr[-1]) + GETJSAMPLE(below_ptr[1]); + neighsum += above_ptr[-1] + above_ptr[1] + below_ptr[-1] + below_ptr[1]; membersum = membersum * memberscale + neighsum * neighscale; *outptr = (JSAMPLE)((membersum + 32768) >> 16); @@ -429,21 +415,18 @@ fullsize_smooth_downsample(j_compress_ptr cinfo, jpeg_component_info *compptr, below_ptr = input_data[outrow + 1]; /* Special case for first column */ - colsum = GETJSAMPLE(*above_ptr++) + GETJSAMPLE(*below_ptr++) + - GETJSAMPLE(*inptr); - membersum = GETJSAMPLE(*inptr++); - nextcolsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(*below_ptr) + - GETJSAMPLE(*inptr); + colsum = (*above_ptr++) + (*below_ptr++) + inptr[0]; + membersum = *inptr++; + nextcolsum = above_ptr[0] + below_ptr[0] + inptr[0]; neighsum = colsum + (colsum - membersum) + nextcolsum; membersum = membersum * memberscale + neighsum * neighscale; *outptr++ = (JSAMPLE)((membersum + 32768) >> 16); lastcolsum = colsum; colsum = nextcolsum; for (colctr = output_cols - 2; colctr > 0; colctr--) { - membersum = GETJSAMPLE(*inptr++); + membersum = *inptr++; above_ptr++; below_ptr++; - nextcolsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(*below_ptr) + - GETJSAMPLE(*inptr); + nextcolsum = above_ptr[0] + below_ptr[0] + inptr[0]; neighsum = lastcolsum + (colsum - membersum) + nextcolsum; membersum = membersum * memberscale + neighsum * neighscale; *outptr++ = (JSAMPLE)((membersum + 32768) >> 16); @@ -451,7 +434,7 @@ fullsize_smooth_downsample(j_compress_ptr cinfo, jpeg_component_info *compptr, } /* Special case for last column */ - membersum = GETJSAMPLE(*inptr); + membersum = *inptr; neighsum = lastcolsum + (colsum - membersum) + colsum; membersum = membersum * memberscale + neighsum * neighscale; *outptr = (JSAMPLE)((membersum + 32768) >> 16); diff --git a/src/3rdparty/libjpeg/src/jctrans.c b/src/3rdparty/libjpeg/src/jctrans.c index ce70a30940..e121028ec7 100644 --- a/src/3rdparty/libjpeg/src/jctrans.c +++ b/src/3rdparty/libjpeg/src/jctrans.c @@ -4,8 +4,8 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1995-1998, Thomas G. Lane. * Modified 2000-2009 by Guido Vollbeding. - * It was modified by The libjpeg-turbo Project to include only code relevant - * to libjpeg-turbo. + * libjpeg-turbo Modifications: + * Copyright (C) 2020, 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -17,6 +17,7 @@ #define JPEG_INTERNALS #include "jinclude.h" #include "jpeglib.h" +#include "jpegcomp.h" /* Forward declarations */ @@ -99,8 +100,8 @@ jpeg_copy_critical_parameters(j_decompress_ptr srcinfo, j_compress_ptr dstinfo) qtblptr = &dstinfo->quant_tbl_ptrs[tblno]; if (*qtblptr == NULL) *qtblptr = jpeg_alloc_quant_table((j_common_ptr)dstinfo); - MEMCOPY((*qtblptr)->quantval, srcinfo->quant_tbl_ptrs[tblno]->quantval, - sizeof((*qtblptr)->quantval)); + memcpy((*qtblptr)->quantval, srcinfo->quant_tbl_ptrs[tblno]->quantval, + sizeof((*qtblptr)->quantval)); (*qtblptr)->sent_table = FALSE; } } diff --git a/src/3rdparty/libjpeg/src/jdapimin.c b/src/3rdparty/libjpeg/src/jdapimin.c index 21a41d2e9f..30126a048d 100644 --- a/src/3rdparty/libjpeg/src/jdapimin.c +++ b/src/3rdparty/libjpeg/src/jdapimin.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1994-1998, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2016, D. R. Commander. + * Copyright (C) 2016, 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -52,7 +52,7 @@ jpeg_CreateDecompress(j_decompress_ptr cinfo, int version, size_t structsize) { struct jpeg_error_mgr *err = cinfo->err; void *client_data = cinfo->client_data; /* ignore Purify complaint here */ - MEMZERO(cinfo, sizeof(struct jpeg_decompress_struct)); + memset(cinfo, 0, sizeof(struct jpeg_decompress_struct)); cinfo->err = err; cinfo->client_data = client_data; } @@ -91,7 +91,7 @@ jpeg_CreateDecompress(j_decompress_ptr cinfo, int version, size_t structsize) cinfo->master = (struct jpeg_decomp_master *) (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_PERMANENT, sizeof(my_decomp_master)); - MEMZERO(cinfo->master, sizeof(my_decomp_master)); + memset(cinfo->master, 0, sizeof(my_decomp_master)); } @@ -308,7 +308,7 @@ jpeg_consume_input(j_decompress_ptr cinfo) /* Initialize application's data source module */ (*cinfo->src->init_source) (cinfo); cinfo->global_state = DSTATE_INHEADER; - /*FALLTHROUGH*/ + FALLTHROUGH /*FALLTHROUGH*/ case DSTATE_INHEADER: retcode = (*cinfo->inputctl->consume_input) (cinfo); if (retcode == JPEG_REACHED_SOS) { /* Found SOS, prepare to decompress */ diff --git a/src/3rdparty/libjpeg/src/jdapistd.c b/src/3rdparty/libjpeg/src/jdapistd.c index 2c808fa564..02cd0cb93a 100644 --- a/src/3rdparty/libjpeg/src/jdapistd.c +++ b/src/3rdparty/libjpeg/src/jdapistd.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1994-1996, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2010, 2015-2018, D. R. Commander. + * Copyright (C) 2010, 2015-2020, 2022, D. R. Commander. * Copyright (C) 2015, Google, Inc. * For conditions of distribution and use, see the accompanying README.ijg * file. @@ -21,6 +21,8 @@ #include "jinclude.h" #include "jdmainct.h" #include "jdcoefct.h" +#include "jdmaster.h" +#include "jdmerge.h" #include "jdsample.h" #include "jmemsys.h" @@ -157,8 +159,12 @@ jpeg_crop_scanline(j_decompress_ptr cinfo, JDIMENSION *xoffset, JDIMENSION input_xoffset; boolean reinit_upsampler = FALSE; jpeg_component_info *compptr; +#ifdef UPSAMPLE_MERGING_SUPPORTED + my_master_ptr master = (my_master_ptr)cinfo->master; +#endif - if (cinfo->global_state != DSTATE_SCANNING || cinfo->output_scanline != 0) + if ((cinfo->global_state != DSTATE_SCANNING && + cinfo->global_state != DSTATE_BUFIMAGE) || cinfo->output_scanline != 0) ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); if (!xoffset || !width) @@ -206,6 +212,13 @@ jpeg_crop_scanline(j_decompress_ptr cinfo, JDIMENSION *xoffset, */ *width = *width + input_xoffset - *xoffset; cinfo->output_width = *width; +#ifdef UPSAMPLE_MERGING_SUPPORTED + if (master->using_merged_upsample && cinfo->max_v_samp_factor == 2) { + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; + upsample->out_row_width = + cinfo->output_width * cinfo->out_color_components; + } +#endif /* Set the first and last iMCU columns that we must decompress. These values * will be used in single-scan decompressions. @@ -316,6 +329,12 @@ LOCAL(void) read_and_discard_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) { JDIMENSION n; +#ifdef UPSAMPLE_MERGING_SUPPORTED + my_master_ptr master = (my_master_ptr)cinfo->master; +#endif + JSAMPLE dummy_sample[1] = { 0 }; + JSAMPROW dummy_row = dummy_sample; + JSAMPARRAY scanlines = NULL; void (*color_convert) (j_decompress_ptr cinfo, JSAMPIMAGE input_buf, JDIMENSION input_row, JSAMPARRAY output_buf, int num_rows) = NULL; @@ -325,6 +344,10 @@ read_and_discard_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) if (cinfo->cconvert && cinfo->cconvert->color_convert) { color_convert = cinfo->cconvert->color_convert; cinfo->cconvert->color_convert = noop_convert; + /* This just prevents UBSan from complaining about adding 0 to a NULL + * pointer. The pointer isn't actually used. + */ + scanlines = &dummy_row; } if (cinfo->cquantize && cinfo->cquantize->color_quantize) { @@ -332,8 +355,15 @@ read_and_discard_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) cinfo->cquantize->color_quantize = noop_quantize; } +#ifdef UPSAMPLE_MERGING_SUPPORTED + if (master->using_merged_upsample && cinfo->max_v_samp_factor == 2) { + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; + scanlines = &upsample->spare_row; + } +#endif + for (n = 0; n < num_lines; n++) - jpeg_read_scanlines(cinfo, NULL, 1); + jpeg_read_scanlines(cinfo, scanlines, 1); if (color_convert) cinfo->cconvert->color_convert = color_convert; @@ -353,6 +383,12 @@ increment_simple_rowgroup_ctr(j_decompress_ptr cinfo, JDIMENSION rows) { JDIMENSION rows_left; my_main_ptr main_ptr = (my_main_ptr)cinfo->main; + my_master_ptr master = (my_master_ptr)cinfo->master; + + if (master->using_merged_upsample && cinfo->max_v_samp_factor == 2) { + read_and_discard_scanlines(cinfo, rows); + return; + } /* Increment the counter to the next row group after the skipped rows. */ main_ptr->rowgroup_ctr += rows / cinfo->max_v_samp_factor; @@ -382,21 +418,27 @@ jpeg_skip_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) { my_main_ptr main_ptr = (my_main_ptr)cinfo->main; my_coef_ptr coef = (my_coef_ptr)cinfo->coef; + my_master_ptr master = (my_master_ptr)cinfo->master; my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; JDIMENSION i, x; int y; JDIMENSION lines_per_iMCU_row, lines_left_in_iMCU_row, lines_after_iMCU_row; JDIMENSION lines_to_skip, lines_to_read; + /* Two-pass color quantization is not supported. */ + if (cinfo->quantize_colors && cinfo->two_pass_quantize) + ERREXIT(cinfo, JERR_NOTIMPL); + if (cinfo->global_state != DSTATE_SCANNING) ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); /* Do not skip past the bottom of the image. */ if (cinfo->output_scanline + num_lines >= cinfo->output_height) { + num_lines = cinfo->output_height - cinfo->output_scanline; cinfo->output_scanline = cinfo->output_height; (*cinfo->inputctl->finish_input_pass) (cinfo); cinfo->inputctl->eoi_reached = TRUE; - return cinfo->output_height - cinfo->output_scanline; + return num_lines; } if (num_lines == 0) @@ -445,8 +487,10 @@ jpeg_skip_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) main_ptr->buffer_full = FALSE; main_ptr->rowgroup_ctr = 0; main_ptr->context_state = CTX_PREPARE_FOR_IMCU; - upsample->next_row_out = cinfo->max_v_samp_factor; - upsample->rows_to_go = cinfo->output_height - cinfo->output_scanline; + if (!master->using_merged_upsample) { + upsample->next_row_out = cinfo->max_v_samp_factor; + upsample->rows_to_go = cinfo->output_height - cinfo->output_scanline; + } } /* Skipping is much simpler when context rows are not required. */ @@ -458,8 +502,10 @@ jpeg_skip_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) cinfo->output_scanline += lines_left_in_iMCU_row; main_ptr->buffer_full = FALSE; main_ptr->rowgroup_ctr = 0; - upsample->next_row_out = cinfo->max_v_samp_factor; - upsample->rows_to_go = cinfo->output_height - cinfo->output_scanline; + if (!master->using_merged_upsample) { + upsample->next_row_out = cinfo->max_v_samp_factor; + upsample->rows_to_go = cinfo->output_height - cinfo->output_scanline; + } } } @@ -480,7 +526,7 @@ jpeg_skip_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) * all of the entropy decoding occurs in jpeg_start_decompress(), assuming * that the input data source is non-suspending. This makes skipping easy. */ - if (cinfo->inputctl->has_multiple_scans) { + if (cinfo->inputctl->has_multiple_scans || cinfo->buffered_image) { if (cinfo->upsample->need_context_rows) { cinfo->output_scanline += lines_to_skip; cinfo->output_iMCU_row += lines_to_skip / lines_per_iMCU_row; @@ -494,7 +540,8 @@ jpeg_skip_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) cinfo->output_iMCU_row += lines_to_skip / lines_per_iMCU_row; increment_simple_rowgroup_ctr(cinfo, lines_to_read); } - upsample->rows_to_go = cinfo->output_height - cinfo->output_scanline; + if (!master->using_merged_upsample) + upsample->rows_to_go = cinfo->output_height - cinfo->output_scanline; return num_lines; } @@ -506,6 +553,8 @@ jpeg_skip_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) * decoded coefficients. This is ~5% faster for large subsets, but * it's tough to tell a difference for smaller images. */ + if (!cinfo->entropy->insufficient_data) + cinfo->master->last_good_iMCU_row = cinfo->input_iMCU_row; (*cinfo->entropy->decode_mcu) (cinfo, NULL); } } @@ -535,7 +584,8 @@ jpeg_skip_scanlines(j_decompress_ptr cinfo, JDIMENSION num_lines) * bit odd, since "rows_to_go" seems to be redundantly keeping track of * output_scanline. */ - upsample->rows_to_go = cinfo->output_height - cinfo->output_scanline; + if (!master->using_merged_upsample) + upsample->rows_to_go = cinfo->output_height - cinfo->output_scanline; /* Always skip the requested number of lines. */ return num_lines; diff --git a/src/3rdparty/libjpeg/src/jdarith.c b/src/3rdparty/libjpeg/src/jdarith.c index 6002481e24..21575e80c7 100644 --- a/src/3rdparty/libjpeg/src/jdarith.c +++ b/src/3rdparty/libjpeg/src/jdarith.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Developed 1997-2015 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2015-2018, D. R. Commander. + * Copyright (C) 2015-2020, 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -80,7 +80,7 @@ get_byte(j_decompress_ptr cinfo) if (!(*src->fill_input_buffer) (cinfo)) ERREXIT(cinfo, JERR_CANT_SUSPEND); src->bytes_in_buffer--; - return GETJOCTET(*src->next_input_byte++); + return *src->next_input_byte++; } @@ -210,13 +210,13 @@ process_restart(j_decompress_ptr cinfo) for (ci = 0; ci < cinfo->comps_in_scan; ci++) { compptr = cinfo->cur_comp_info[ci]; if (!cinfo->progressive_mode || (cinfo->Ss == 0 && cinfo->Ah == 0)) { - MEMZERO(entropy->dc_stats[compptr->dc_tbl_no], DC_STAT_BINS); + memset(entropy->dc_stats[compptr->dc_tbl_no], 0, DC_STAT_BINS); /* Reset DC predictions to 0 */ entropy->last_dc_val[ci] = 0; entropy->dc_context[ci] = 0; } if (!cinfo->progressive_mode || cinfo->Ss) { - MEMZERO(entropy->ac_stats[compptr->ac_tbl_no], AC_STAT_BINS); + memset(entropy->ac_stats[compptr->ac_tbl_no], 0, AC_STAT_BINS); } } @@ -471,17 +471,17 @@ decode_mcu_AC_refine(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) if (*thiscoef) { /* previously nonzero coef */ if (arith_decode(cinfo, st + 2)) { if (*thiscoef < 0) - *thiscoef += m1; + *thiscoef += (JCOEF)m1; else - *thiscoef += p1; + *thiscoef += (JCOEF)p1; } break; } if (arith_decode(cinfo, st + 1)) { /* newly nonzero coef */ if (arith_decode(cinfo, entropy->fixed_bin)) - *thiscoef = m1; + *thiscoef = (JCOEF)m1; else - *thiscoef = p1; + *thiscoef = (JCOEF)p1; break; } st += 3; k++; @@ -665,8 +665,16 @@ bad: for (ci = 0; ci < cinfo->comps_in_scan; ci++) { int coefi, cindex = cinfo->cur_comp_info[ci]->component_index; int *coef_bit_ptr = &cinfo->coef_bits[cindex][0]; + int *prev_coef_bit_ptr = + &cinfo->coef_bits[cindex + cinfo->num_components][0]; if (cinfo->Ss && coef_bit_ptr[0] < 0) /* AC without prior DC scan */ WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, 0); + for (coefi = MIN(cinfo->Ss, 1); coefi <= MAX(cinfo->Se, 9); coefi++) { + if (cinfo->input_scan_number > 1) + prev_coef_bit_ptr[coefi] = coef_bit_ptr[coefi]; + else + prev_coef_bit_ptr[coefi] = 0; + } for (coefi = cinfo->Ss; coefi <= cinfo->Se; coefi++) { int expected = (coef_bit_ptr[coefi] < 0) ? 0 : coef_bit_ptr[coefi]; if (cinfo->Ah != expected) @@ -690,8 +698,8 @@ bad: /* Check that the scan parameters Ss, Se, Ah/Al are OK for sequential JPEG. * This ought to be an error condition, but we make it a warning. */ - if (cinfo->Ss != 0 || cinfo->Ah != 0 || cinfo->Al != 0 || - (cinfo->Se < DCTSIZE2 && cinfo->Se != DCTSIZE2 - 1)) + if (cinfo->Ss != 0 || cinfo->Se != DCTSIZE2 - 1 || + cinfo->Ah != 0 || cinfo->Al != 0) WARNMS(cinfo, JWRN_NOT_SEQUENTIAL); /* Select MCU decoding routine */ entropy->pub.decode_mcu = decode_mcu; @@ -707,7 +715,7 @@ bad: if (entropy->dc_stats[tbl] == NULL) entropy->dc_stats[tbl] = (unsigned char *)(*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, DC_STAT_BINS); - MEMZERO(entropy->dc_stats[tbl], DC_STAT_BINS); + memset(entropy->dc_stats[tbl], 0, DC_STAT_BINS); /* Initialize DC predictions to 0 */ entropy->last_dc_val[ci] = 0; entropy->dc_context[ci] = 0; @@ -719,7 +727,7 @@ bad: if (entropy->ac_stats[tbl] == NULL) entropy->ac_stats[tbl] = (unsigned char *)(*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, AC_STAT_BINS); - MEMZERO(entropy->ac_stats[tbl], AC_STAT_BINS); + memset(entropy->ac_stats[tbl], 0, AC_STAT_BINS); } } @@ -727,6 +735,7 @@ bad: entropy->c = 0; entropy->a = 0; entropy->ct = -16; /* force reading 2 initial bytes to fill C */ + entropy->pub.insufficient_data = FALSE; /* Initialize restart counter */ entropy->restarts_to_go = cinfo->restart_interval; @@ -763,7 +772,7 @@ jinit_arith_decoder(j_decompress_ptr cinfo) int *coef_bit_ptr, ci; cinfo->coef_bits = (int (*)[DCTSIZE2]) (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, - cinfo->num_components * DCTSIZE2 * + cinfo->num_components * 2 * DCTSIZE2 * sizeof(int)); coef_bit_ptr = &cinfo->coef_bits[0][0]; for (ci = 0; ci < cinfo->num_components; ci++) diff --git a/src/3rdparty/libjpeg/src/jdatadst.c b/src/3rdparty/libjpeg/src/jdatadst.c index 246fffb58a..6b4fed2339 100644 --- a/src/3rdparty/libjpeg/src/jdatadst.c +++ b/src/3rdparty/libjpeg/src/jdatadst.c @@ -5,7 +5,7 @@ * Copyright (C) 1994-1996, Thomas G. Lane. * Modified 2009-2012 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2013, 2016, D. R. Commander. + * Copyright (C) 2013, 2016, 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -23,11 +23,6 @@ #include "jpeglib.h" #include "jerror.h" -#ifndef HAVE_STDLIB_H /* <stdlib.h> should declare malloc(),free() */ -extern void *malloc(size_t size); -extern void free(void *ptr); -#endif - /* Expanded data destination object for stdio output */ @@ -116,7 +111,7 @@ empty_output_buffer(j_compress_ptr cinfo) { my_dest_ptr dest = (my_dest_ptr)cinfo->dest; - if (JFWRITE(dest->outfile, dest->buffer, OUTPUT_BUF_SIZE) != + if (fwrite(dest->buffer, 1, OUTPUT_BUF_SIZE, dest->outfile) != (size_t)OUTPUT_BUF_SIZE) ERREXIT(cinfo, JERR_FILE_WRITE); @@ -141,7 +136,7 @@ empty_mem_output_buffer(j_compress_ptr cinfo) if (nextbuffer == NULL) ERREXIT1(cinfo, JERR_OUT_OF_MEMORY, 10); - MEMCOPY(nextbuffer, dest->buffer, dest->bufsize); + memcpy(nextbuffer, dest->buffer, dest->bufsize); free(dest->newbuffer); @@ -175,7 +170,7 @@ term_destination(j_compress_ptr cinfo) /* Write any data remaining in the buffer */ if (datacount > 0) { - if (JFWRITE(dest->outfile, dest->buffer, datacount) != datacount) + if (fwrite(dest->buffer, 1, datacount, dest->outfile) != datacount) ERREXIT(cinfo, JERR_FILE_WRITE); } fflush(dest->outfile); diff --git a/src/3rdparty/libjpeg/src/jdatasrc.c b/src/3rdparty/libjpeg/src/jdatasrc.c index eadb4a2c90..e36a30d894 100644 --- a/src/3rdparty/libjpeg/src/jdatasrc.c +++ b/src/3rdparty/libjpeg/src/jdatasrc.c @@ -5,7 +5,7 @@ * Copyright (C) 1994-1996, Thomas G. Lane. * Modified 2009-2011 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2013, 2016, D. R. Commander. + * Copyright (C) 2013, 2016, 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -104,7 +104,7 @@ fill_input_buffer(j_decompress_ptr cinfo) my_src_ptr src = (my_src_ptr)cinfo->src; size_t nbytes; - nbytes = JFREAD(src->infile, src->buffer, INPUT_BUF_SIZE); + nbytes = fread(src->buffer, 1, INPUT_BUF_SIZE, src->infile); if (nbytes <= 0) { if (src->start_of_file) /* Treat empty input file as fatal error */ diff --git a/src/3rdparty/libjpeg/src/jdcoefct.c b/src/3rdparty/libjpeg/src/jdcoefct.c index 723a9ac2be..88e10c08cb 100644 --- a/src/3rdparty/libjpeg/src/jdcoefct.c +++ b/src/3rdparty/libjpeg/src/jdcoefct.c @@ -5,8 +5,8 @@ * Copyright (C) 1994-1997, Thomas G. Lane. * libjpeg-turbo Modifications: * Copyright 2009 Pierre Ossman <ossman@cendio.se> for Cendio AB - * Copyright (C) 2010, 2015-2016, D. R. Commander. - * Copyright (C) 2015, Google, Inc. + * Copyright (C) 2010, 2015-2016, 2019-2020, 2022, D. R. Commander. + * Copyright (C) 2015, 2020, Google, Inc. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -102,6 +102,8 @@ decompress_onepass(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) /* Try to fetch an MCU. Entropy decoder expects buffer to be zeroed. */ jzero_far((void *)coef->MCU_buffer[0], (size_t)(cinfo->blocks_in_MCU * sizeof(JBLOCK))); + if (!cinfo->entropy->insufficient_data) + cinfo->master->last_good_iMCU_row = cinfo->input_iMCU_row; if (!(*cinfo->entropy->decode_mcu) (cinfo, coef->MCU_buffer)) { /* Suspension forced; update state counters and exit */ coef->MCU_vert_offset = yoffset; @@ -227,6 +229,8 @@ consume_data(j_decompress_ptr cinfo) } } } + if (!cinfo->entropy->insufficient_data) + cinfo->master->last_good_iMCU_row = cinfo->input_iMCU_row; /* Try to fetch the MCU. */ if (!(*cinfo->entropy->decode_mcu) (cinfo, coef->MCU_buffer)) { /* Suspension forced; update state counters and exit */ @@ -326,19 +330,22 @@ decompress_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) #ifdef BLOCK_SMOOTHING_SUPPORTED /* - * This code applies interblock smoothing as described by section K.8 - * of the JPEG standard: the first 5 AC coefficients are estimated from - * the DC values of a DCT block and its 8 neighboring blocks. + * This code applies interblock smoothing; the first 9 AC coefficients are + * estimated from the DC values of a DCT block and its 24 neighboring blocks. * We apply smoothing only for progressive JPEG decoding, and only if * the coefficients it can estimate are not yet known to full precision. */ -/* Natural-order array positions of the first 5 zigzag-order coefficients */ +/* Natural-order array positions of the first 9 zigzag-order coefficients */ #define Q01_POS 1 #define Q10_POS 8 #define Q20_POS 16 #define Q11_POS 9 #define Q02_POS 2 +#define Q03_POS 3 +#define Q12_POS 10 +#define Q21_POS 17 +#define Q30_POS 24 /* * Determine whether block smoothing is applicable and safe. @@ -356,8 +363,8 @@ smoothing_ok(j_decompress_ptr cinfo) int ci, coefi; jpeg_component_info *compptr; JQUANT_TBL *qtable; - int *coef_bits; - int *coef_bits_latch; + int *coef_bits, *prev_coef_bits; + int *coef_bits_latch, *prev_coef_bits_latch; if (!cinfo->progressive_mode || cinfo->coef_bits == NULL) return FALSE; @@ -366,34 +373,47 @@ smoothing_ok(j_decompress_ptr cinfo) if (coef->coef_bits_latch == NULL) coef->coef_bits_latch = (int *) (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, - cinfo->num_components * + cinfo->num_components * 2 * (SAVED_COEFS * sizeof(int))); coef_bits_latch = coef->coef_bits_latch; + prev_coef_bits_latch = + &coef->coef_bits_latch[cinfo->num_components * SAVED_COEFS]; for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; ci++, compptr++) { /* All components' quantization values must already be latched. */ if ((qtable = compptr->quant_table) == NULL) return FALSE; - /* Verify DC & first 5 AC quantizers are nonzero to avoid zero-divide. */ + /* Verify DC & first 9 AC quantizers are nonzero to avoid zero-divide. */ if (qtable->quantval[0] == 0 || qtable->quantval[Q01_POS] == 0 || qtable->quantval[Q10_POS] == 0 || qtable->quantval[Q20_POS] == 0 || qtable->quantval[Q11_POS] == 0 || - qtable->quantval[Q02_POS] == 0) + qtable->quantval[Q02_POS] == 0 || + qtable->quantval[Q03_POS] == 0 || + qtable->quantval[Q12_POS] == 0 || + qtable->quantval[Q21_POS] == 0 || + qtable->quantval[Q30_POS] == 0) return FALSE; /* DC values must be at least partly known for all components. */ coef_bits = cinfo->coef_bits[ci]; + prev_coef_bits = cinfo->coef_bits[ci + cinfo->num_components]; if (coef_bits[0] < 0) return FALSE; + coef_bits_latch[0] = coef_bits[0]; /* Block smoothing is helpful if some AC coefficients remain inaccurate. */ - for (coefi = 1; coefi <= 5; coefi++) { + for (coefi = 1; coefi < SAVED_COEFS; coefi++) { + if (cinfo->input_scan_number > 1) + prev_coef_bits_latch[coefi] = prev_coef_bits[coefi]; + else + prev_coef_bits_latch[coefi] = -1; coef_bits_latch[coefi] = coef_bits[coefi]; if (coef_bits[coefi] != 0) smoothing_useful = TRUE; } coef_bits_latch += SAVED_COEFS; + prev_coef_bits_latch += SAVED_COEFS; } return smoothing_useful; @@ -412,17 +432,20 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) JDIMENSION block_num, last_block_column; int ci, block_row, block_rows, access_rows; JBLOCKARRAY buffer; - JBLOCKROW buffer_ptr, prev_block_row, next_block_row; + JBLOCKROW buffer_ptr, prev_prev_block_row, prev_block_row; + JBLOCKROW next_block_row, next_next_block_row; JSAMPARRAY output_ptr; JDIMENSION output_col; jpeg_component_info *compptr; inverse_DCT_method_ptr inverse_DCT; - boolean first_row, last_row; + boolean change_dc; JCOEF *workspace; int *coef_bits; JQUANT_TBL *quanttbl; - JLONG Q00, Q01, Q02, Q10, Q11, Q20, num; - int DC1, DC2, DC3, DC4, DC5, DC6, DC7, DC8, DC9; + JLONG Q00, Q01, Q02, Q03 = 0, Q10, Q11, Q12 = 0, Q20, Q21 = 0, Q30 = 0, num; + int DC01, DC02, DC03, DC04, DC05, DC06, DC07, DC08, DC09, DC10, DC11, DC12, + DC13, DC14, DC15, DC16, DC17, DC18, DC19, DC20, DC21, DC22, DC23, DC24, + DC25; int Al, pred; /* Keep a local variable to avoid looking it up more than once */ @@ -434,10 +457,10 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) if (cinfo->input_scan_number == cinfo->output_scan_number) { /* If input is working on current scan, we ordinarily want it to * have completed the current row. But if input scan is DC, - * we want it to keep one row ahead so that next block row's DC + * we want it to keep two rows ahead so that next two block rows' DC * values are up to date. */ - JDIMENSION delta = (cinfo->Ss == 0) ? 1 : 0; + JDIMENSION delta = (cinfo->Ss == 0) ? 2 : 0; if (cinfo->input_iMCU_row > cinfo->output_iMCU_row + delta) break; } @@ -452,34 +475,53 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) if (!compptr->component_needed) continue; /* Count non-dummy DCT block rows in this iMCU row. */ - if (cinfo->output_iMCU_row < last_iMCU_row) { + if (cinfo->output_iMCU_row + 1 < last_iMCU_row) { + block_rows = compptr->v_samp_factor; + access_rows = block_rows * 3; /* this and next two iMCU rows */ + } else if (cinfo->output_iMCU_row < last_iMCU_row) { block_rows = compptr->v_samp_factor; access_rows = block_rows * 2; /* this and next iMCU row */ - last_row = FALSE; } else { /* NB: can't use last_row_height here; it is input-side-dependent! */ block_rows = (int)(compptr->height_in_blocks % compptr->v_samp_factor); if (block_rows == 0) block_rows = compptr->v_samp_factor; access_rows = block_rows; /* this iMCU row only */ - last_row = TRUE; } /* Align the virtual buffer for this component. */ - if (cinfo->output_iMCU_row > 0) { - access_rows += compptr->v_samp_factor; /* prior iMCU row too */ + if (cinfo->output_iMCU_row > 1) { + access_rows += 2 * compptr->v_samp_factor; /* prior two iMCU rows too */ + buffer = (*cinfo->mem->access_virt_barray) + ((j_common_ptr)cinfo, coef->whole_image[ci], + (cinfo->output_iMCU_row - 2) * compptr->v_samp_factor, + (JDIMENSION)access_rows, FALSE); + buffer += 2 * compptr->v_samp_factor; /* point to current iMCU row */ + } else if (cinfo->output_iMCU_row > 0) { buffer = (*cinfo->mem->access_virt_barray) ((j_common_ptr)cinfo, coef->whole_image[ci], (cinfo->output_iMCU_row - 1) * compptr->v_samp_factor, (JDIMENSION)access_rows, FALSE); buffer += compptr->v_samp_factor; /* point to current iMCU row */ - first_row = FALSE; } else { buffer = (*cinfo->mem->access_virt_barray) ((j_common_ptr)cinfo, coef->whole_image[ci], (JDIMENSION)0, (JDIMENSION)access_rows, FALSE); - first_row = TRUE; } - /* Fetch component-dependent info */ - coef_bits = coef->coef_bits_latch + (ci * SAVED_COEFS); + /* Fetch component-dependent info. + * If the current scan is incomplete, then we use the component-dependent + * info from the previous scan. + */ + if (cinfo->output_iMCU_row > cinfo->master->last_good_iMCU_row) + coef_bits = + coef->coef_bits_latch + ((ci + cinfo->num_components) * SAVED_COEFS); + else + coef_bits = coef->coef_bits_latch + (ci * SAVED_COEFS); + + /* We only do DC interpolation if no AC coefficient data is available. */ + change_dc = + coef_bits[1] == -1 && coef_bits[2] == -1 && coef_bits[3] == -1 && + coef_bits[4] == -1 && coef_bits[5] == -1 && coef_bits[6] == -1 && + coef_bits[7] == -1 && coef_bits[8] == -1 && coef_bits[9] == -1; + quanttbl = compptr->quant_table; Q00 = quanttbl->quantval[0]; Q01 = quanttbl->quantval[Q01_POS]; @@ -487,25 +529,51 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) Q20 = quanttbl->quantval[Q20_POS]; Q11 = quanttbl->quantval[Q11_POS]; Q02 = quanttbl->quantval[Q02_POS]; + if (change_dc) { + Q03 = quanttbl->quantval[Q03_POS]; + Q12 = quanttbl->quantval[Q12_POS]; + Q21 = quanttbl->quantval[Q21_POS]; + Q30 = quanttbl->quantval[Q30_POS]; + } inverse_DCT = cinfo->idct->inverse_DCT[ci]; output_ptr = output_buf[ci]; /* Loop over all DCT blocks to be processed. */ for (block_row = 0; block_row < block_rows; block_row++) { buffer_ptr = buffer[block_row] + cinfo->master->first_MCU_col[ci]; - if (first_row && block_row == 0) + + if (block_row > 0 || cinfo->output_iMCU_row > 0) + prev_block_row = + buffer[block_row - 1] + cinfo->master->first_MCU_col[ci]; + else prev_block_row = buffer_ptr; + + if (block_row > 1 || cinfo->output_iMCU_row > 1) + prev_prev_block_row = + buffer[block_row - 2] + cinfo->master->first_MCU_col[ci]; + else + prev_prev_block_row = prev_block_row; + + if (block_row < block_rows - 1 || cinfo->output_iMCU_row < last_iMCU_row) + next_block_row = + buffer[block_row + 1] + cinfo->master->first_MCU_col[ci]; else - prev_block_row = buffer[block_row - 1]; - if (last_row && block_row == block_rows - 1) next_block_row = buffer_ptr; + + if (block_row < block_rows - 2 || + cinfo->output_iMCU_row + 1 < last_iMCU_row) + next_next_block_row = + buffer[block_row + 2] + cinfo->master->first_MCU_col[ci]; else - next_block_row = buffer[block_row + 1]; + next_next_block_row = next_block_row; + /* We fetch the surrounding DC values using a sliding-register approach. - * Initialize all nine here so as to do the right thing on narrow pics. + * Initialize all 25 here so as to do the right thing on narrow pics. */ - DC1 = DC2 = DC3 = (int)prev_block_row[0][0]; - DC4 = DC5 = DC6 = (int)buffer_ptr[0][0]; - DC7 = DC8 = DC9 = (int)next_block_row[0][0]; + DC01 = DC02 = DC03 = DC04 = DC05 = (int)prev_prev_block_row[0][0]; + DC06 = DC07 = DC08 = DC09 = DC10 = (int)prev_block_row[0][0]; + DC11 = DC12 = DC13 = DC14 = DC15 = (int)buffer_ptr[0][0]; + DC16 = DC17 = DC18 = DC19 = DC20 = (int)next_block_row[0][0]; + DC21 = DC22 = DC23 = DC24 = DC25 = (int)next_next_block_row[0][0]; output_col = 0; last_block_column = compptr->width_in_blocks - 1; for (block_num = cinfo->master->first_MCU_col[ci]; @@ -513,18 +581,39 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) /* Fetch current DCT block into workspace so we can modify it. */ jcopy_block_row(buffer_ptr, (JBLOCKROW)workspace, (JDIMENSION)1); /* Update DC values */ - if (block_num < last_block_column) { - DC3 = (int)prev_block_row[1][0]; - DC6 = (int)buffer_ptr[1][0]; - DC9 = (int)next_block_row[1][0]; + if (block_num == cinfo->master->first_MCU_col[ci] && + block_num < last_block_column) { + DC04 = (int)prev_prev_block_row[1][0]; + DC09 = (int)prev_block_row[1][0]; + DC14 = (int)buffer_ptr[1][0]; + DC19 = (int)next_block_row[1][0]; + DC24 = (int)next_next_block_row[1][0]; } - /* Compute coefficient estimates per K.8. - * An estimate is applied only if coefficient is still zero, - * and is not known to be fully accurate. + if (block_num + 1 < last_block_column) { + DC05 = (int)prev_prev_block_row[2][0]; + DC10 = (int)prev_block_row[2][0]; + DC15 = (int)buffer_ptr[2][0]; + DC20 = (int)next_block_row[2][0]; + DC25 = (int)next_next_block_row[2][0]; + } + /* If DC interpolation is enabled, compute coefficient estimates using + * a Gaussian-like kernel, keeping the averages of the DC values. + * + * If DC interpolation is disabled, compute coefficient estimates using + * an algorithm similar to the one described in Section K.8 of the JPEG + * standard, except applied to a 5x5 window rather than a 3x3 window. + * + * An estimate is applied only if the coefficient is still zero and is + * not known to be fully accurate. */ /* AC01 */ if ((Al = coef_bits[1]) != 0 && workspace[1] == 0) { - num = 36 * Q00 * (DC4 - DC6); + num = Q00 * (change_dc ? + (-DC01 - DC02 + DC04 + DC05 - 3 * DC06 + 13 * DC07 - + 13 * DC09 + 3 * DC10 - 3 * DC11 + 38 * DC12 - 38 * DC14 + + 3 * DC15 - 3 * DC16 + 13 * DC17 - 13 * DC19 + 3 * DC20 - + DC21 - DC22 + DC24 + DC25) : + (-7 * DC11 + 50 * DC12 - 50 * DC14 + 7 * DC15)); if (num >= 0) { pred = (int)(((Q01 << 7) + num) / (Q01 << 8)); if (Al > 0 && pred >= (1 << Al)) @@ -539,7 +628,12 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) } /* AC10 */ if ((Al = coef_bits[2]) != 0 && workspace[8] == 0) { - num = 36 * Q00 * (DC2 - DC8); + num = Q00 * (change_dc ? + (-DC01 - 3 * DC02 - 3 * DC03 - 3 * DC04 - DC05 - DC06 + + 13 * DC07 + 38 * DC08 + 13 * DC09 - DC10 + DC16 - + 13 * DC17 - 38 * DC18 - 13 * DC19 + DC20 + DC21 + + 3 * DC22 + 3 * DC23 + 3 * DC24 + DC25) : + (-7 * DC03 + 50 * DC08 - 50 * DC18 + 7 * DC23)); if (num >= 0) { pred = (int)(((Q10 << 7) + num) / (Q10 << 8)); if (Al > 0 && pred >= (1 << Al)) @@ -554,7 +648,10 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) } /* AC20 */ if ((Al = coef_bits[3]) != 0 && workspace[16] == 0) { - num = 9 * Q00 * (DC2 + DC8 - 2 * DC5); + num = Q00 * (change_dc ? + (DC03 + 2 * DC07 + 7 * DC08 + 2 * DC09 - 5 * DC12 - 14 * DC13 - + 5 * DC14 + 2 * DC17 + 7 * DC18 + 2 * DC19 + DC23) : + (-DC03 + 13 * DC08 - 24 * DC13 + 13 * DC18 - DC23)); if (num >= 0) { pred = (int)(((Q20 << 7) + num) / (Q20 << 8)); if (Al > 0 && pred >= (1 << Al)) @@ -569,7 +666,11 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) } /* AC11 */ if ((Al = coef_bits[4]) != 0 && workspace[9] == 0) { - num = 5 * Q00 * (DC1 - DC3 - DC7 + DC9); + num = Q00 * (change_dc ? + (-DC01 + DC05 + 9 * DC07 - 9 * DC09 - 9 * DC17 + + 9 * DC19 + DC21 - DC25) : + (DC10 + DC16 - 10 * DC17 + 10 * DC19 - DC02 - DC20 + DC22 - + DC24 + DC04 - DC06 + 10 * DC07 - 10 * DC09)); if (num >= 0) { pred = (int)(((Q11 << 7) + num) / (Q11 << 8)); if (Al > 0 && pred >= (1 << Al)) @@ -584,7 +685,10 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) } /* AC02 */ if ((Al = coef_bits[5]) != 0 && workspace[2] == 0) { - num = 9 * Q00 * (DC4 + DC6 - 2 * DC5); + num = Q00 * (change_dc ? + (2 * DC07 - 5 * DC08 + 2 * DC09 + DC11 + 7 * DC12 - 14 * DC13 + + 7 * DC14 + DC15 + 2 * DC17 - 5 * DC18 + 2 * DC19) : + (-DC11 + 13 * DC12 - 24 * DC13 + 13 * DC14 - DC15)); if (num >= 0) { pred = (int)(((Q02 << 7) + num) / (Q02 << 8)); if (Al > 0 && pred >= (1 << Al)) @@ -597,14 +701,96 @@ decompress_smooth_data(j_decompress_ptr cinfo, JSAMPIMAGE output_buf) } workspace[2] = (JCOEF)pred; } + if (change_dc) { + /* AC03 */ + if ((Al = coef_bits[6]) != 0 && workspace[3] == 0) { + num = Q00 * (DC07 - DC09 + 2 * DC12 - 2 * DC14 + DC17 - DC19); + if (num >= 0) { + pred = (int)(((Q03 << 7) + num) / (Q03 << 8)); + if (Al > 0 && pred >= (1 << Al)) + pred = (1 << Al) - 1; + } else { + pred = (int)(((Q03 << 7) - num) / (Q03 << 8)); + if (Al > 0 && pred >= (1 << Al)) + pred = (1 << Al) - 1; + pred = -pred; + } + workspace[3] = (JCOEF)pred; + } + /* AC12 */ + if ((Al = coef_bits[7]) != 0 && workspace[10] == 0) { + num = Q00 * (DC07 - 3 * DC08 + DC09 - DC17 + 3 * DC18 - DC19); + if (num >= 0) { + pred = (int)(((Q12 << 7) + num) / (Q12 << 8)); + if (Al > 0 && pred >= (1 << Al)) + pred = (1 << Al) - 1; + } else { + pred = (int)(((Q12 << 7) - num) / (Q12 << 8)); + if (Al > 0 && pred >= (1 << Al)) + pred = (1 << Al) - 1; + pred = -pred; + } + workspace[10] = (JCOEF)pred; + } + /* AC21 */ + if ((Al = coef_bits[8]) != 0 && workspace[17] == 0) { + num = Q00 * (DC07 - DC09 - 3 * DC12 + 3 * DC14 + DC17 - DC19); + if (num >= 0) { + pred = (int)(((Q21 << 7) + num) / (Q21 << 8)); + if (Al > 0 && pred >= (1 << Al)) + pred = (1 << Al) - 1; + } else { + pred = (int)(((Q21 << 7) - num) / (Q21 << 8)); + if (Al > 0 && pred >= (1 << Al)) + pred = (1 << Al) - 1; + pred = -pred; + } + workspace[17] = (JCOEF)pred; + } + /* AC30 */ + if ((Al = coef_bits[9]) != 0 && workspace[24] == 0) { + num = Q00 * (DC07 + 2 * DC08 + DC09 - DC17 - 2 * DC18 - DC19); + if (num >= 0) { + pred = (int)(((Q30 << 7) + num) / (Q30 << 8)); + if (Al > 0 && pred >= (1 << Al)) + pred = (1 << Al) - 1; + } else { + pred = (int)(((Q30 << 7) - num) / (Q30 << 8)); + if (Al > 0 && pred >= (1 << Al)) + pred = (1 << Al) - 1; + pred = -pred; + } + workspace[24] = (JCOEF)pred; + } + /* coef_bits[0] is non-negative. Otherwise this function would not + * be called. + */ + num = Q00 * + (-2 * DC01 - 6 * DC02 - 8 * DC03 - 6 * DC04 - 2 * DC05 - + 6 * DC06 + 6 * DC07 + 42 * DC08 + 6 * DC09 - 6 * DC10 - + 8 * DC11 + 42 * DC12 + 152 * DC13 + 42 * DC14 - 8 * DC15 - + 6 * DC16 + 6 * DC17 + 42 * DC18 + 6 * DC19 - 6 * DC20 - + 2 * DC21 - 6 * DC22 - 8 * DC23 - 6 * DC24 - 2 * DC25); + if (num >= 0) { + pred = (int)(((Q00 << 7) + num) / (Q00 << 8)); + } else { + pred = (int)(((Q00 << 7) - num) / (Q00 << 8)); + pred = -pred; + } + workspace[0] = (JCOEF)pred; + } /* change_dc */ + /* OK, do the IDCT */ (*inverse_DCT) (cinfo, compptr, (JCOEFPTR)workspace, output_ptr, output_col); /* Advance for next column */ - DC1 = DC2; DC2 = DC3; - DC4 = DC5; DC5 = DC6; - DC7 = DC8; DC8 = DC9; - buffer_ptr++, prev_block_row++, next_block_row++; + DC01 = DC02; DC02 = DC03; DC03 = DC04; DC04 = DC05; + DC06 = DC07; DC07 = DC08; DC08 = DC09; DC09 = DC10; + DC11 = DC12; DC12 = DC13; DC13 = DC14; DC14 = DC15; + DC16 = DC17; DC17 = DC18; DC18 = DC19; DC19 = DC20; + DC21 = DC22; DC22 = DC23; DC23 = DC24; DC24 = DC25; + buffer_ptr++, prev_block_row++, next_block_row++, + prev_prev_block_row++, next_next_block_row++; output_col += compptr->_DCT_scaled_size; } output_ptr += compptr->_DCT_scaled_size; @@ -653,7 +839,7 @@ jinit_d_coef_controller(j_decompress_ptr cinfo, boolean need_full_buffer) #ifdef BLOCK_SMOOTHING_SUPPORTED /* If block smoothing could be used, need a bigger window */ if (cinfo->progressive_mode) - access_rows *= 3; + access_rows *= 5; #endif coef->whole_image[ci] = (*cinfo->mem->request_virt_barray) ((j_common_ptr)cinfo, JPOOL_IMAGE, TRUE, diff --git a/src/3rdparty/libjpeg/src/jdcoefct.h b/src/3rdparty/libjpeg/src/jdcoefct.h index c4d1943dd4..9a0e780663 100644 --- a/src/3rdparty/libjpeg/src/jdcoefct.h +++ b/src/3rdparty/libjpeg/src/jdcoefct.h @@ -5,6 +5,7 @@ * Copyright (C) 1994-1997, Thomas G. Lane. * libjpeg-turbo Modifications: * Copyright 2009 Pierre Ossman <ossman@cendio.se> for Cendio AB + * Copyright (C) 2020, Google, Inc. * For conditions of distribution and use, see the accompanying README.ijg * file. */ @@ -51,7 +52,7 @@ typedef struct { #ifdef BLOCK_SMOOTHING_SUPPORTED /* When doing block smoothing, we latch coefficient Al values here */ int *coef_bits_latch; -#define SAVED_COEFS 6 /* we save coef_bits[0..5] */ +#define SAVED_COEFS 10 /* we save coef_bits[0..9] */ #endif } my_coef_controller; diff --git a/src/3rdparty/libjpeg/src/jdcol565.c b/src/3rdparty/libjpeg/src/jdcol565.c index 40068ef84f..53c7bd9187 100644 --- a/src/3rdparty/libjpeg/src/jdcol565.c +++ b/src/3rdparty/libjpeg/src/jdcol565.c @@ -45,9 +45,9 @@ ycc_rgb565_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, outptr = *output_buf++; if (PACK_NEED_ALIGNMENT(outptr)) { - y = GETJSAMPLE(*inptr0++); - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + y = *inptr0++; + cb = *inptr1++; + cr = *inptr2++; r = range_limit[y + Crrtab[cr]]; g = range_limit[y + ((int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS))]; @@ -58,18 +58,18 @@ ycc_rgb565_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, num_cols--; } for (col = 0; col < (num_cols >> 1); col++) { - y = GETJSAMPLE(*inptr0++); - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + y = *inptr0++; + cb = *inptr1++; + cr = *inptr2++; r = range_limit[y + Crrtab[cr]]; g = range_limit[y + ((int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS))]; b = range_limit[y + Cbbtab[cb]]; rgb = PACK_SHORT_565(r, g, b); - y = GETJSAMPLE(*inptr0++); - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + y = *inptr0++; + cb = *inptr1++; + cr = *inptr2++; r = range_limit[y + Crrtab[cr]]; g = range_limit[y + ((int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS))]; @@ -80,9 +80,9 @@ ycc_rgb565_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, outptr += 4; } if (num_cols & 1) { - y = GETJSAMPLE(*inptr0); - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); + y = *inptr0; + cb = *inptr1; + cr = *inptr2; r = range_limit[y + Crrtab[cr]]; g = range_limit[y + ((int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS))]; @@ -125,9 +125,9 @@ ycc_rgb565D_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, input_row++; outptr = *output_buf++; if (PACK_NEED_ALIGNMENT(outptr)) { - y = GETJSAMPLE(*inptr0++); - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + y = *inptr0++; + cb = *inptr1++; + cr = *inptr2++; r = range_limit[DITHER_565_R(y + Crrtab[cr], d0)]; g = range_limit[DITHER_565_G(y + ((int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], @@ -139,9 +139,9 @@ ycc_rgb565D_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, num_cols--; } for (col = 0; col < (num_cols >> 1); col++) { - y = GETJSAMPLE(*inptr0++); - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + y = *inptr0++; + cb = *inptr1++; + cr = *inptr2++; r = range_limit[DITHER_565_R(y + Crrtab[cr], d0)]; g = range_limit[DITHER_565_G(y + ((int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], @@ -150,9 +150,9 @@ ycc_rgb565D_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, d0 = DITHER_ROTATE(d0); rgb = PACK_SHORT_565(r, g, b); - y = GETJSAMPLE(*inptr0++); - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + y = *inptr0++; + cb = *inptr1++; + cr = *inptr2++; r = range_limit[DITHER_565_R(y + Crrtab[cr], d0)]; g = range_limit[DITHER_565_G(y + ((int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], @@ -165,9 +165,9 @@ ycc_rgb565D_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, outptr += 4; } if (num_cols & 1) { - y = GETJSAMPLE(*inptr0); - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); + y = *inptr0; + cb = *inptr1; + cr = *inptr2; r = range_limit[DITHER_565_R(y + Crrtab[cr], d0)]; g = range_limit[DITHER_565_G(y + ((int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], @@ -202,32 +202,32 @@ rgb_rgb565_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, input_row++; outptr = *output_buf++; if (PACK_NEED_ALIGNMENT(outptr)) { - r = GETJSAMPLE(*inptr0++); - g = GETJSAMPLE(*inptr1++); - b = GETJSAMPLE(*inptr2++); + r = *inptr0++; + g = *inptr1++; + b = *inptr2++; rgb = PACK_SHORT_565(r, g, b); *(INT16 *)outptr = (INT16)rgb; outptr += 2; num_cols--; } for (col = 0; col < (num_cols >> 1); col++) { - r = GETJSAMPLE(*inptr0++); - g = GETJSAMPLE(*inptr1++); - b = GETJSAMPLE(*inptr2++); + r = *inptr0++; + g = *inptr1++; + b = *inptr2++; rgb = PACK_SHORT_565(r, g, b); - r = GETJSAMPLE(*inptr0++); - g = GETJSAMPLE(*inptr1++); - b = GETJSAMPLE(*inptr2++); + r = *inptr0++; + g = *inptr1++; + b = *inptr2++; rgb = PACK_TWO_PIXELS(rgb, PACK_SHORT_565(r, g, b)); WRITE_TWO_ALIGNED_PIXELS(outptr, rgb); outptr += 4; } if (num_cols & 1) { - r = GETJSAMPLE(*inptr0); - g = GETJSAMPLE(*inptr1); - b = GETJSAMPLE(*inptr2); + r = *inptr0; + g = *inptr1; + b = *inptr2; rgb = PACK_SHORT_565(r, g, b); *(INT16 *)outptr = (INT16)rgb; } @@ -259,24 +259,24 @@ rgb_rgb565D_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, input_row++; outptr = *output_buf++; if (PACK_NEED_ALIGNMENT(outptr)) { - r = range_limit[DITHER_565_R(GETJSAMPLE(*inptr0++), d0)]; - g = range_limit[DITHER_565_G(GETJSAMPLE(*inptr1++), d0)]; - b = range_limit[DITHER_565_B(GETJSAMPLE(*inptr2++), d0)]; + r = range_limit[DITHER_565_R(*inptr0++, d0)]; + g = range_limit[DITHER_565_G(*inptr1++, d0)]; + b = range_limit[DITHER_565_B(*inptr2++, d0)]; rgb = PACK_SHORT_565(r, g, b); *(INT16 *)outptr = (INT16)rgb; outptr += 2; num_cols--; } for (col = 0; col < (num_cols >> 1); col++) { - r = range_limit[DITHER_565_R(GETJSAMPLE(*inptr0++), d0)]; - g = range_limit[DITHER_565_G(GETJSAMPLE(*inptr1++), d0)]; - b = range_limit[DITHER_565_B(GETJSAMPLE(*inptr2++), d0)]; + r = range_limit[DITHER_565_R(*inptr0++, d0)]; + g = range_limit[DITHER_565_G(*inptr1++, d0)]; + b = range_limit[DITHER_565_B(*inptr2++, d0)]; d0 = DITHER_ROTATE(d0); rgb = PACK_SHORT_565(r, g, b); - r = range_limit[DITHER_565_R(GETJSAMPLE(*inptr0++), d0)]; - g = range_limit[DITHER_565_G(GETJSAMPLE(*inptr1++), d0)]; - b = range_limit[DITHER_565_B(GETJSAMPLE(*inptr2++), d0)]; + r = range_limit[DITHER_565_R(*inptr0++, d0)]; + g = range_limit[DITHER_565_G(*inptr1++, d0)]; + b = range_limit[DITHER_565_B(*inptr2++, d0)]; d0 = DITHER_ROTATE(d0); rgb = PACK_TWO_PIXELS(rgb, PACK_SHORT_565(r, g, b)); @@ -284,9 +284,9 @@ rgb_rgb565D_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, outptr += 4; } if (num_cols & 1) { - r = range_limit[DITHER_565_R(GETJSAMPLE(*inptr0), d0)]; - g = range_limit[DITHER_565_G(GETJSAMPLE(*inptr1), d0)]; - b = range_limit[DITHER_565_B(GETJSAMPLE(*inptr2), d0)]; + r = range_limit[DITHER_565_R(*inptr0, d0)]; + g = range_limit[DITHER_565_G(*inptr1, d0)]; + b = range_limit[DITHER_565_B(*inptr2, d0)]; rgb = PACK_SHORT_565(r, g, b); *(INT16 *)outptr = (INT16)rgb; } diff --git a/src/3rdparty/libjpeg/src/jdcolext.c b/src/3rdparty/libjpeg/src/jdcolext.c index 72a5301070..fc7e7b8f00 100644 --- a/src/3rdparty/libjpeg/src/jdcolext.c +++ b/src/3rdparty/libjpeg/src/jdcolext.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1997, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2009, 2011, 2015, D. R. Commander. + * Copyright (C) 2009, 2011, 2015, 2023, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -53,19 +53,19 @@ ycc_rgb_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, input_row++; outptr = *output_buf++; for (col = 0; col < num_cols; col++) { - y = GETJSAMPLE(inptr0[col]); - cb = GETJSAMPLE(inptr1[col]); - cr = GETJSAMPLE(inptr2[col]); + y = inptr0[col]; + cb = inptr1[col]; + cr = inptr2[col]; /* Range-limiting is essential due to noise introduced by DCT losses. */ outptr[RGB_RED] = range_limit[y + Crrtab[cr]]; outptr[RGB_GREEN] = range_limit[y + ((int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS))]; outptr[RGB_BLUE] = range_limit[y + Cbbtab[cb]]; - /* Set unused byte to 0xFF so it can be interpreted as an opaque */ + /* Set unused byte to MAXJSAMPLE so it can be interpreted as an opaque */ /* alpha channel value */ #ifdef RGB_ALPHA - outptr[RGB_ALPHA] = 0xFF; + outptr[RGB_ALPHA] = MAXJSAMPLE; #endif outptr += RGB_PIXELSIZE; } @@ -93,12 +93,11 @@ gray_rgb_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, inptr = input_buf[0][input_row++]; outptr = *output_buf++; for (col = 0; col < num_cols; col++) { - /* We can dispense with GETJSAMPLE() here */ outptr[RGB_RED] = outptr[RGB_GREEN] = outptr[RGB_BLUE] = inptr[col]; - /* Set unused byte to 0xFF so it can be interpreted as an opaque */ + /* Set unused byte to MAXJSAMPLE so it can be interpreted as an opaque */ /* alpha channel value */ #ifdef RGB_ALPHA - outptr[RGB_ALPHA] = 0xFF; + outptr[RGB_ALPHA] = MAXJSAMPLE; #endif outptr += RGB_PIXELSIZE; } @@ -128,14 +127,13 @@ rgb_rgb_convert_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, input_row++; outptr = *output_buf++; for (col = 0; col < num_cols; col++) { - /* We can dispense with GETJSAMPLE() here */ outptr[RGB_RED] = inptr0[col]; outptr[RGB_GREEN] = inptr1[col]; outptr[RGB_BLUE] = inptr2[col]; - /* Set unused byte to 0xFF so it can be interpreted as an opaque */ + /* Set unused byte to MAXJSAMPLE so it can be interpreted as an opaque */ /* alpha channel value */ #ifdef RGB_ALPHA - outptr[RGB_ALPHA] = 0xFF; + outptr[RGB_ALPHA] = MAXJSAMPLE; #endif outptr += RGB_PIXELSIZE; } diff --git a/src/3rdparty/libjpeg/src/jdcolor.c b/src/3rdparty/libjpeg/src/jdcolor.c index dc0e3b6c0e..735190b700 100644 --- a/src/3rdparty/libjpeg/src/jdcolor.c +++ b/src/3rdparty/libjpeg/src/jdcolor.c @@ -18,7 +18,6 @@ #include "jinclude.h" #include "jpeglib.h" #include "jsimd.h" -#include "jconfigint.h" /* Private subobject */ @@ -341,9 +340,9 @@ rgb_gray_convert(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, input_row++; outptr = *output_buf++; for (col = 0; col < num_cols; col++) { - r = GETJSAMPLE(inptr0[col]); - g = GETJSAMPLE(inptr1[col]); - b = GETJSAMPLE(inptr2[col]); + r = inptr0[col]; + g = inptr1[col]; + b = inptr2[col]; /* Y */ outptr[col] = (JSAMPLE)((ctab[r + R_Y_OFF] + ctab[g + G_Y_OFF] + ctab[b + B_Y_OFF]) >> SCALEBITS); @@ -550,9 +549,9 @@ ycck_cmyk_convert(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, input_row++; outptr = *output_buf++; for (col = 0; col < num_cols; col++) { - y = GETJSAMPLE(inptr0[col]); - cb = GETJSAMPLE(inptr1[col]); - cr = GETJSAMPLE(inptr2[col]); + y = inptr0[col]; + cb = inptr1[col]; + cr = inptr2[col]; /* Range-limiting is essential due to noise introduced by DCT losses. */ outptr[0] = range_limit[MAXJSAMPLE - (y + Crrtab[cr])]; /* red */ outptr[1] = range_limit[MAXJSAMPLE - (y + /* green */ @@ -560,7 +559,7 @@ ycck_cmyk_convert(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, SCALEBITS)))]; outptr[2] = range_limit[MAXJSAMPLE - (y + Cbbtab[cb])]; /* blue */ /* K passes through unchanged */ - outptr[3] = inptr3[col]; /* don't need GETJSAMPLE here */ + outptr[3] = inptr3[col]; outptr += 4; } } @@ -571,11 +570,10 @@ ycck_cmyk_convert(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, * RGB565 conversion */ -#define PACK_SHORT_565_LE(r, g, b) ((((r) << 8) & 0xF800) | \ - (((g) << 3) & 0x7E0) | ((b) >> 3)) -#define PACK_SHORT_565_BE(r, g, b) (((r) & 0xF8) | ((g) >> 5) | \ - (((g) << 11) & 0xE000) | \ - (((b) << 5) & 0x1F00)) +#define PACK_SHORT_565_LE(r, g, b) \ + ((((r) << 8) & 0xF800) | (((g) << 3) & 0x7E0) | ((b) >> 3)) +#define PACK_SHORT_565_BE(r, g, b) \ + (((r) & 0xF8) | ((g) >> 5) | (((g) << 11) & 0xE000) | (((b) << 5) & 0x1F00)) #define PACK_TWO_PIXELS_LE(l, r) ((r << 16) | l) #define PACK_TWO_PIXELS_BE(l, r) ((l << 16) | r) diff --git a/src/3rdparty/libjpeg/src/jddctmgr.c b/src/3rdparty/libjpeg/src/jddctmgr.c index 266f446623..e78d7bebe2 100644 --- a/src/3rdparty/libjpeg/src/jddctmgr.c +++ b/src/3rdparty/libjpeg/src/jddctmgr.c @@ -6,7 +6,7 @@ * Modified 2002-2010 by Guido Vollbeding. * libjpeg-turbo Modifications: * Copyright 2009 Pierre Ossman <ossman@cendio.se> for Cendio AB - * Copyright (C) 2010, 2015, D. R. Commander. + * Copyright (C) 2010, 2015, 2022, D. R. Commander. * Copyright (C) 2013, MIPS Technologies, Inc., California. * For conditions of distribution and use, see the accompanying README.ijg * file. @@ -345,7 +345,7 @@ jinit_inverse_dct(j_decompress_ptr cinfo) compptr->dct_table = (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, sizeof(multiplier_table)); - MEMZERO(compptr->dct_table, sizeof(multiplier_table)); + memset(compptr->dct_table, 0, sizeof(multiplier_table)); /* Mark multiplier table not yet set up for any method */ idct->cur_method[ci] = -1; } diff --git a/src/3rdparty/libjpeg/src/jdhuff.c b/src/3rdparty/libjpeg/src/jdhuff.c index a1128178b0..679d221685 100644 --- a/src/3rdparty/libjpeg/src/jdhuff.c +++ b/src/3rdparty/libjpeg/src/jdhuff.c @@ -5,6 +5,7 @@ * Copyright (C) 1991-1997, Thomas G. Lane. * libjpeg-turbo Modifications: * Copyright (C) 2009-2011, 2016, 2018-2019, D. R. Commander. + * Copyright (C) 2018, Matthias Räncker. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -39,24 +40,6 @@ typedef struct { int last_dc_val[MAX_COMPS_IN_SCAN]; /* last DC coef for each component */ } savable_state; -/* This macro is to work around compilers with missing or broken - * structure assignment. You'll need to fix this code if you have - * such a compiler and you change MAX_COMPS_IN_SCAN. - */ - -#ifndef NO_STRUCT_ASSIGN -#define ASSIGN_STATE(dest, src) ((dest) = (src)) -#else -#if MAX_COMPS_IN_SCAN == 4 -#define ASSIGN_STATE(dest, src) \ - ((dest).last_dc_val[0] = (src).last_dc_val[0], \ - (dest).last_dc_val[1] = (src).last_dc_val[1], \ - (dest).last_dc_val[2] = (src).last_dc_val[2], \ - (dest).last_dc_val[3] = (src).last_dc_val[3]) -#endif -#endif - - typedef struct { struct jpeg_entropy_decoder pub; /* public fields */ @@ -325,7 +308,7 @@ jpeg_fill_bit_buffer(bitread_working_state *state, bytes_in_buffer = cinfo->src->bytes_in_buffer; } bytes_in_buffer--; - c = GETJOCTET(*next_input_byte++); + c = *next_input_byte++; /* If it's 0xFF, check and discard stuffed zero byte */ if (c == 0xFF) { @@ -342,7 +325,7 @@ jpeg_fill_bit_buffer(bitread_working_state *state, bytes_in_buffer = cinfo->src->bytes_in_buffer; } bytes_in_buffer--; - c = GETJOCTET(*next_input_byte++); + c = *next_input_byte++; } while (c == 0xFF); if (c == 0) { @@ -405,8 +388,8 @@ no_more_bytes: #define GET_BYTE { \ register int c0, c1; \ - c0 = GETJOCTET(*buffer++); \ - c1 = GETJOCTET(*buffer); \ + c0 = *buffer++; \ + c1 = *buffer; \ /* Pre-execute most common case */ \ get_buffer = (get_buffer << 8) | c0; \ bits_left += 8; \ @@ -423,7 +406,7 @@ no_more_bytes: } \ } -#if SIZEOF_SIZE_T == 8 || defined(_WIN64) +#if SIZEOF_SIZE_T == 8 || defined(_WIN64) || (defined(__x86_64__) && defined(__ILP32__)) /* Pre-fetch 48 bytes, because the holding register is 64-bit */ #define FILL_BIT_BUFFER_FAST \ @@ -557,6 +540,12 @@ process_restart(j_decompress_ptr cinfo) } +#if defined(__has_feature) +#if __has_feature(undefined_behavior_sanitizer) +__attribute__((no_sanitize("signed-integer-overflow"), + no_sanitize("unsigned-integer-overflow"))) +#endif +#endif LOCAL(boolean) decode_mcu_slow(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) { @@ -568,7 +557,7 @@ decode_mcu_slow(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) /* Load up working state */ BITREAD_LOAD_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(state, entropy->saved); + state = entropy->saved; for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { JBLOCKROW block = MCU_data ? MCU_data[blkn] : NULL; @@ -589,11 +578,15 @@ decode_mcu_slow(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) if (entropy->dc_needed[blkn]) { /* Convert DC difference to actual value, update last_dc_val */ int ci = cinfo->MCU_membership[blkn]; - /* This is really just - * s += state.last_dc_val[ci]; - * It is written this way in order to shut up UBSan. + /* Certain malformed JPEG images produce repeated DC coefficient + * differences of 2047 or -2047, which causes state.last_dc_val[ci] to + * grow until it overflows or underflows a 32-bit signed integer. This + * behavior is, to the best of our understanding, innocuous, and it is + * unclear how to work around it without potentially affecting + * performance. Thus, we (hopefully temporarily) suppress UBSan integer + * overflow errors for this function and decode_mcu_fast(). */ - s = (int)((unsigned int)s + (unsigned int)state.last_dc_val[ci]); + s += state.last_dc_val[ci]; state.last_dc_val[ci] = s; if (block) { /* Output the DC coefficient (assumes jpeg_natural_order[0] = 0) */ @@ -653,11 +646,17 @@ decode_mcu_slow(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) /* Completed MCU, so update state */ BITREAD_SAVE_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(entropy->saved, state); + entropy->saved = state; return TRUE; } +#if defined(__has_feature) +#if __has_feature(undefined_behavior_sanitizer) +__attribute__((no_sanitize("signed-integer-overflow"), + no_sanitize("unsigned-integer-overflow"))) +#endif +#endif LOCAL(boolean) decode_mcu_fast(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) { @@ -671,7 +670,7 @@ decode_mcu_fast(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) /* Load up working state */ BITREAD_LOAD_STATE(cinfo, entropy->bitstate); buffer = (JOCTET *)br_state.next_input_byte; - ASSIGN_STATE(state, entropy->saved); + state = entropy->saved; for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { JBLOCKROW block = MCU_data ? MCU_data[blkn] : NULL; @@ -688,7 +687,10 @@ decode_mcu_fast(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) if (entropy->dc_needed[blkn]) { int ci = cinfo->MCU_membership[blkn]; - s = (int)((unsigned int)s + (unsigned int)state.last_dc_val[ci]); + /* Refer to the comment in decode_mcu_slow() regarding the supression of + * a UBSan integer overflow error in this line of code. + */ + s += state.last_dc_val[ci]; state.last_dc_val[ci] = s; if (block) (*block)[0] = (JCOEF)s; @@ -740,7 +742,7 @@ decode_mcu_fast(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) br_state.bytes_in_buffer -= (buffer - br_state.next_input_byte); br_state.next_input_byte = buffer; BITREAD_SAVE_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(entropy->saved, state); + entropy->saved = state; return TRUE; } @@ -795,7 +797,8 @@ use_slow: } /* Account for restart interval (no-op if not using restarts) */ - entropy->restarts_to_go--; + if (cinfo->restart_interval) + entropy->restarts_to_go--; return TRUE; } diff --git a/src/3rdparty/libjpeg/src/jdhuff.h b/src/3rdparty/libjpeg/src/jdhuff.h index 6a8d90f402..cfa0b7f558 100644 --- a/src/3rdparty/libjpeg/src/jdhuff.h +++ b/src/3rdparty/libjpeg/src/jdhuff.h @@ -4,7 +4,8 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1997, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2010-2011, 2015-2016, D. R. Commander. + * Copyright (C) 2010-2011, 2015-2016, 2021, D. R. Commander. + * Copyright (C) 2018, Matthias Räncker. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -78,6 +79,11 @@ EXTERN(void) jpeg_make_d_derived_tbl(j_decompress_ptr cinfo, boolean isDC, typedef size_t bit_buf_type; /* type of bit-extraction buffer */ #define BIT_BUF_SIZE 64 /* size of buffer in bits */ +#elif defined(__x86_64__) && defined(__ILP32__) + +typedef unsigned long long bit_buf_type; /* type of bit-extraction buffer */ +#define BIT_BUF_SIZE 64 /* size of buffer in bits */ + #else typedef unsigned long bit_buf_type; /* type of bit-extraction buffer */ @@ -228,7 +234,10 @@ slowlabel: \ s |= GET_BITS(1); \ nb++; \ } \ - s = htbl->pub->huffval[(int)(s + htbl->valoffset[nb]) & 0xFF]; \ + if (nb > 16) \ + s = 0; \ + else \ + s = htbl->pub->huffval[(int)(s + htbl->valoffset[nb]) & 0xFF]; \ } /* Out-of-line case for Huffman code fetching */ diff --git a/src/3rdparty/libjpeg/src/jdinput.c b/src/3rdparty/libjpeg/src/jdinput.c index deec618f26..1bc5aff1a7 100644 --- a/src/3rdparty/libjpeg/src/jdinput.c +++ b/src/3rdparty/libjpeg/src/jdinput.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1997, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2010, 2016, 2018, D. R. Commander. + * Copyright (C) 2010, 2016, 2018, 2022, D. R. Commander. * Copyright (C) 2015, Google, Inc. * For conditions of distribution and use, see the accompanying README.ijg * file. @@ -264,7 +264,7 @@ latch_quant_tables(j_decompress_ptr cinfo) qtbl = (JQUANT_TBL *) (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, sizeof(JQUANT_TBL)); - MEMCOPY(qtbl, cinfo->quant_tbl_ptrs[qtblno], sizeof(JQUANT_TBL)); + memcpy(qtbl, cinfo->quant_tbl_ptrs[qtblno], sizeof(JQUANT_TBL)); compptr->quant_table = qtbl; } } diff --git a/src/3rdparty/libjpeg/src/jdmainct.c b/src/3rdparty/libjpeg/src/jdmainct.c index 50301d6b50..d332e6b2fa 100644 --- a/src/3rdparty/libjpeg/src/jdmainct.c +++ b/src/3rdparty/libjpeg/src/jdmainct.c @@ -360,7 +360,7 @@ process_data_context_main(j_decompress_ptr cinfo, JSAMPARRAY output_buf, main_ptr->context_state = CTX_PREPARE_FOR_IMCU; if (*out_row_ctr >= out_rows_avail) return; /* Postprocessor exactly filled output buf */ - /*FALLTHROUGH*/ + FALLTHROUGH /*FALLTHROUGH*/ case CTX_PREPARE_FOR_IMCU: /* Prepare to process first M-1 row groups of this iMCU row */ main_ptr->rowgroup_ctr = 0; @@ -371,7 +371,7 @@ process_data_context_main(j_decompress_ptr cinfo, JSAMPARRAY output_buf, if (main_ptr->iMCU_row_ctr == cinfo->total_iMCU_rows) set_bottom_pointers(cinfo); main_ptr->context_state = CTX_PROCESS_IMCU; - /*FALLTHROUGH*/ + FALLTHROUGH /*FALLTHROUGH*/ case CTX_PROCESS_IMCU: /* Call postprocessor using previously set pointers */ (*cinfo->post->post_process_data) (cinfo, diff --git a/src/3rdparty/libjpeg/src/jdmarker.c b/src/3rdparty/libjpeg/src/jdmarker.c index c9c7ef6399..f7eba615fd 100644 --- a/src/3rdparty/libjpeg/src/jdmarker.c +++ b/src/3rdparty/libjpeg/src/jdmarker.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1998, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2012, 2015, D. R. Commander. + * Copyright (C) 2012, 2015, 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -151,7 +151,7 @@ typedef my_marker_reader *my_marker_ptr; #define INPUT_BYTE(cinfo, V, action) \ MAKESTMT( MAKE_BYTE_AVAIL(cinfo, action); \ bytes_in_buffer--; \ - V = GETJOCTET(*next_input_byte++); ) + V = *next_input_byte++; ) /* As above, but read two bytes interpreted as an unsigned 16-bit integer. * V should be declared unsigned int or perhaps JLONG. @@ -159,10 +159,10 @@ typedef my_marker_reader *my_marker_ptr; #define INPUT_2BYTES(cinfo, V, action) \ MAKESTMT( MAKE_BYTE_AVAIL(cinfo, action); \ bytes_in_buffer--; \ - V = ((unsigned int)GETJOCTET(*next_input_byte++)) << 8; \ + V = ((unsigned int)(*next_input_byte++)) << 8; \ MAKE_BYTE_AVAIL(cinfo, action); \ bytes_in_buffer--; \ - V += GETJOCTET(*next_input_byte++); ) + V += *next_input_byte++; ) /* @@ -473,7 +473,7 @@ get_dht(j_decompress_ptr cinfo) for (i = 0; i < count; i++) INPUT_BYTE(cinfo, huffval[i], return FALSE); - MEMZERO(&huffval[count], (256 - count) * sizeof(UINT8)); + memset(&huffval[count], 0, (256 - count) * sizeof(UINT8)); length -= count; @@ -491,8 +491,8 @@ get_dht(j_decompress_ptr cinfo) if (*htblptr == NULL) *htblptr = jpeg_alloc_huff_table((j_common_ptr)cinfo); - MEMCOPY((*htblptr)->bits, bits, sizeof((*htblptr)->bits)); - MEMCOPY((*htblptr)->huffval, huffval, sizeof((*htblptr)->huffval)); + memcpy((*htblptr)->bits, bits, sizeof((*htblptr)->bits)); + memcpy((*htblptr)->huffval, huffval, sizeof((*htblptr)->huffval)); } if (length != 0) @@ -608,18 +608,18 @@ examine_app0(j_decompress_ptr cinfo, JOCTET *data, unsigned int datalen, JLONG totallen = (JLONG)datalen + remaining; if (datalen >= APP0_DATA_LEN && - GETJOCTET(data[0]) == 0x4A && - GETJOCTET(data[1]) == 0x46 && - GETJOCTET(data[2]) == 0x49 && - GETJOCTET(data[3]) == 0x46 && - GETJOCTET(data[4]) == 0) { + data[0] == 0x4A && + data[1] == 0x46 && + data[2] == 0x49 && + data[3] == 0x46 && + data[4] == 0) { /* Found JFIF APP0 marker: save info */ cinfo->saw_JFIF_marker = TRUE; - cinfo->JFIF_major_version = GETJOCTET(data[5]); - cinfo->JFIF_minor_version = GETJOCTET(data[6]); - cinfo->density_unit = GETJOCTET(data[7]); - cinfo->X_density = (GETJOCTET(data[8]) << 8) + GETJOCTET(data[9]); - cinfo->Y_density = (GETJOCTET(data[10]) << 8) + GETJOCTET(data[11]); + cinfo->JFIF_major_version = data[5]; + cinfo->JFIF_minor_version = data[6]; + cinfo->density_unit = data[7]; + cinfo->X_density = (data[8] << 8) + data[9]; + cinfo->Y_density = (data[10] << 8) + data[11]; /* Check version. * Major version must be 1, anything else signals an incompatible change. * (We used to treat this as an error, but now it's a nonfatal warning, @@ -634,24 +634,22 @@ examine_app0(j_decompress_ptr cinfo, JOCTET *data, unsigned int datalen, cinfo->JFIF_major_version, cinfo->JFIF_minor_version, cinfo->X_density, cinfo->Y_density, cinfo->density_unit); /* Validate thumbnail dimensions and issue appropriate messages */ - if (GETJOCTET(data[12]) | GETJOCTET(data[13])) - TRACEMS2(cinfo, 1, JTRC_JFIF_THUMBNAIL, - GETJOCTET(data[12]), GETJOCTET(data[13])); + if (data[12] | data[13]) + TRACEMS2(cinfo, 1, JTRC_JFIF_THUMBNAIL, data[12], data[13]); totallen -= APP0_DATA_LEN; - if (totallen != - ((JLONG)GETJOCTET(data[12]) * (JLONG)GETJOCTET(data[13]) * (JLONG)3)) + if (totallen != ((JLONG)data[12] * (JLONG)data[13] * (JLONG)3)) TRACEMS1(cinfo, 1, JTRC_JFIF_BADTHUMBNAILSIZE, (int)totallen); } else if (datalen >= 6 && - GETJOCTET(data[0]) == 0x4A && - GETJOCTET(data[1]) == 0x46 && - GETJOCTET(data[2]) == 0x58 && - GETJOCTET(data[3]) == 0x58 && - GETJOCTET(data[4]) == 0) { + data[0] == 0x4A && + data[1] == 0x46 && + data[2] == 0x58 && + data[3] == 0x58 && + data[4] == 0) { /* Found JFIF "JFXX" extension APP0 marker */ /* The library doesn't actually do anything with these, * but we try to produce a helpful trace message. */ - switch (GETJOCTET(data[5])) { + switch (data[5]) { case 0x10: TRACEMS1(cinfo, 1, JTRC_THUMB_JPEG, (int)totallen); break; @@ -662,8 +660,7 @@ examine_app0(j_decompress_ptr cinfo, JOCTET *data, unsigned int datalen, TRACEMS1(cinfo, 1, JTRC_THUMB_RGB, (int)totallen); break; default: - TRACEMS2(cinfo, 1, JTRC_JFIF_EXTENSION, - GETJOCTET(data[5]), (int)totallen); + TRACEMS2(cinfo, 1, JTRC_JFIF_EXTENSION, data[5], (int)totallen); break; } } else { @@ -684,16 +681,16 @@ examine_app14(j_decompress_ptr cinfo, JOCTET *data, unsigned int datalen, unsigned int version, flags0, flags1, transform; if (datalen >= APP14_DATA_LEN && - GETJOCTET(data[0]) == 0x41 && - GETJOCTET(data[1]) == 0x64 && - GETJOCTET(data[2]) == 0x6F && - GETJOCTET(data[3]) == 0x62 && - GETJOCTET(data[4]) == 0x65) { + data[0] == 0x41 && + data[1] == 0x64 && + data[2] == 0x6F && + data[3] == 0x62 && + data[4] == 0x65) { /* Found Adobe APP14 marker */ - version = (GETJOCTET(data[5]) << 8) + GETJOCTET(data[6]); - flags0 = (GETJOCTET(data[7]) << 8) + GETJOCTET(data[8]); - flags1 = (GETJOCTET(data[9]) << 8) + GETJOCTET(data[10]); - transform = GETJOCTET(data[11]); + version = (data[5] << 8) + data[6]; + flags0 = (data[7] << 8) + data[8]; + flags1 = (data[9] << 8) + data[10]; + transform = data[11]; TRACEMS4(cinfo, 1, JTRC_ADOBE, version, flags0, flags1, transform); cinfo->saw_Adobe_marker = TRUE; cinfo->Adobe_transform = (UINT8)transform; diff --git a/src/3rdparty/libjpeg/src/jdmaster.c b/src/3rdparty/libjpeg/src/jdmaster.c index b20906438e..a3690bf560 100644 --- a/src/3rdparty/libjpeg/src/jdmaster.c +++ b/src/3rdparty/libjpeg/src/jdmaster.c @@ -5,7 +5,7 @@ * Copyright (C) 1991-1997, Thomas G. Lane. * Modified 2002-2009 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2009-2011, 2016, D. R. Commander. + * Copyright (C) 2009-2011, 2016, 2019, 2022, D. R. Commander. * Copyright (C) 2013, Linaro Limited. * Copyright (C) 2015, Google, Inc. * For conditions of distribution and use, see the accompanying README.ijg @@ -22,7 +22,6 @@ #include "jpeglib.h" #include "jpegcomp.h" #include "jdmaster.h" -#include "jsimd.h" /* @@ -70,17 +69,6 @@ use_merged_upsample(j_decompress_ptr cinfo) cinfo->comp_info[1]._DCT_scaled_size != cinfo->_min_DCT_scaled_size || cinfo->comp_info[2]._DCT_scaled_size != cinfo->_min_DCT_scaled_size) return FALSE; -#ifdef WITH_SIMD - /* If YCbCr-to-RGB color conversion is SIMD-accelerated but merged upsampling - isn't, then disabling merged upsampling is likely to be faster when - decompressing YCbCr JPEG images. */ - if (!jsimd_can_h2v2_merged_upsample() && !jsimd_can_h2v1_merged_upsample() && - jsimd_can_ycc_rgb() && cinfo->jpeg_color_space == JCS_YCbCr && - (cinfo->out_color_space == JCS_RGB || - (cinfo->out_color_space >= JCS_EXT_RGB && - cinfo->out_color_space <= JCS_EXT_ARGB))) - return FALSE; -#endif /* ??? also need to test for upsample-time rescaling, when & if supported */ return TRUE; /* by golly, it'll work... */ #else @@ -429,7 +417,7 @@ prepare_range_limit_table(j_decompress_ptr cinfo) table += (MAXJSAMPLE + 1); /* allow negative subscripts of simple table */ cinfo->sample_range_limit = table; /* First segment of "simple" table: limit[x] = 0 for x < 0 */ - MEMZERO(table - (MAXJSAMPLE + 1), (MAXJSAMPLE + 1) * sizeof(JSAMPLE)); + memset(table - (MAXJSAMPLE + 1), 0, (MAXJSAMPLE + 1) * sizeof(JSAMPLE)); /* Main part of "simple" table: limit[x] = x */ for (i = 0; i <= MAXJSAMPLE; i++) table[i] = (JSAMPLE)i; @@ -438,10 +426,10 @@ prepare_range_limit_table(j_decompress_ptr cinfo) for (i = CENTERJSAMPLE; i < 2 * (MAXJSAMPLE + 1); i++) table[i] = MAXJSAMPLE; /* Second half of post-IDCT table */ - MEMZERO(table + (2 * (MAXJSAMPLE + 1)), - (2 * (MAXJSAMPLE + 1) - CENTERJSAMPLE) * sizeof(JSAMPLE)); - MEMCOPY(table + (4 * (MAXJSAMPLE + 1) - CENTERJSAMPLE), - cinfo->sample_range_limit, CENTERJSAMPLE * sizeof(JSAMPLE)); + memset(table + (2 * (MAXJSAMPLE + 1)), 0, + (2 * (MAXJSAMPLE + 1) - CENTERJSAMPLE) * sizeof(JSAMPLE)); + memcpy(table + (4 * (MAXJSAMPLE + 1) - CENTERJSAMPLE), + cinfo->sample_range_limit, CENTERJSAMPLE * sizeof(JSAMPLE)); } @@ -580,6 +568,7 @@ master_selection(j_decompress_ptr cinfo) */ cinfo->master->first_iMCU_col = 0; cinfo->master->last_iMCU_col = cinfo->MCUs_per_row - 1; + cinfo->master->last_good_iMCU_row = 0; #ifdef D_MULTISCAN_FILES_SUPPORTED /* If jpeg_start_decompress will read the whole file, initialize diff --git a/src/3rdparty/libjpeg/src/jdmerge.c b/src/3rdparty/libjpeg/src/jdmerge.c index dff5a35087..38b002729c 100644 --- a/src/3rdparty/libjpeg/src/jdmerge.c +++ b/src/3rdparty/libjpeg/src/jdmerge.c @@ -5,7 +5,7 @@ * Copyright (C) 1994-1996, Thomas G. Lane. * libjpeg-turbo Modifications: * Copyright 2009 Pierre Ossman <ossman@cendio.se> for Cendio AB - * Copyright (C) 2009, 2011, 2014-2015, D. R. Commander. + * Copyright (C) 2009, 2011, 2014-2015, 2020, D. R. Commander. * Copyright (C) 2013, Linaro Limited. * For conditions of distribution and use, see the accompanying README.ijg * file. @@ -40,41 +40,12 @@ #define JPEG_INTERNALS #include "jinclude.h" #include "jpeglib.h" +#include "jdmerge.h" #include "jsimd.h" -#include "jconfigint.h" #ifdef UPSAMPLE_MERGING_SUPPORTED -/* Private subobject */ - -typedef struct { - struct jpeg_upsampler pub; /* public fields */ - - /* Pointer to routine to do actual upsampling/conversion of one row group */ - void (*upmethod) (j_decompress_ptr cinfo, JSAMPIMAGE input_buf, - JDIMENSION in_row_group_ctr, JSAMPARRAY output_buf); - - /* Private state for YCC->RGB conversion */ - int *Cr_r_tab; /* => table for Cr to R conversion */ - int *Cb_b_tab; /* => table for Cb to B conversion */ - JLONG *Cr_g_tab; /* => table for Cr to G conversion */ - JLONG *Cb_g_tab; /* => table for Cb to G conversion */ - - /* For 2:1 vertical sampling, we produce two output rows at a time. - * We need a "spare" row buffer to hold the second output row if the - * application provides just a one-row buffer; we also use the spare - * to discard the dummy last row if the image height is odd. - */ - JSAMPROW spare_row; - boolean spare_full; /* T if spare buffer is occupied */ - - JDIMENSION out_row_width; /* samples per output row */ - JDIMENSION rows_to_go; /* counts rows remaining in image */ -} my_upsampler; - -typedef my_upsampler *my_upsample_ptr; - #define SCALEBITS 16 /* speediest right-shift on some machines */ #define ONE_HALF ((JLONG)1 << (SCALEBITS - 1)) #define FIX(x) ((JLONG)((x) * (1L << SCALEBITS) + 0.5)) @@ -189,7 +160,7 @@ typedef my_upsampler *my_upsample_ptr; LOCAL(void) build_ycc_rgb_table(j_decompress_ptr cinfo) { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; int i; JLONG x; SHIFT_TEMPS @@ -232,7 +203,7 @@ build_ycc_rgb_table(j_decompress_ptr cinfo) METHODDEF(void) start_pass_merged_upsample(j_decompress_ptr cinfo) { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; /* Mark the spare buffer empty */ upsample->spare_full = FALSE; @@ -254,7 +225,7 @@ merged_2v_upsample(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail) /* 2:1 vertical sampling case: may need a spare row. */ { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; JSAMPROW work_ptrs[2]; JDIMENSION num_rows; /* number of rows returned to caller */ @@ -305,7 +276,7 @@ merged_1v_upsample(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail) /* 1:1 vertical sampling case: much easier, never need a spare row. */ { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; /* Just do the upsampling. */ (*upsample->upmethod) (cinfo, input_buf, *in_row_group_ctr, @@ -420,11 +391,10 @@ h2v2_merged_upsample(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, * RGB565 conversion */ -#define PACK_SHORT_565_LE(r, g, b) ((((r) << 8) & 0xF800) | \ - (((g) << 3) & 0x7E0) | ((b) >> 3)) -#define PACK_SHORT_565_BE(r, g, b) (((r) & 0xF8) | ((g) >> 5) | \ - (((g) << 11) & 0xE000) | \ - (((b) << 5) & 0x1F00)) +#define PACK_SHORT_565_LE(r, g, b) \ + ((((r) << 8) & 0xF800) | (((g) << 3) & 0x7E0) | ((b) >> 3)) +#define PACK_SHORT_565_BE(r, g, b) \ + (((r) & 0xF8) | ((g) >> 5) | (((g) << 11) & 0xE000) | (((b) << 5) & 0x1F00)) #define PACK_TWO_PIXELS_LE(l, r) ((r << 16) | l) #define PACK_TWO_PIXELS_BE(l, r) ((l << 16) | r) @@ -566,11 +536,11 @@ h2v2_merged_upsample_565D(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, GLOBAL(void) jinit_merged_upsampler(j_decompress_ptr cinfo) { - my_upsample_ptr upsample; + my_merged_upsample_ptr upsample; - upsample = (my_upsample_ptr) + upsample = (my_merged_upsample_ptr) (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, - sizeof(my_upsampler)); + sizeof(my_merged_upsampler)); cinfo->upsample = (struct jpeg_upsampler *)upsample; upsample->pub.start_pass = start_pass_merged_upsample; upsample->pub.need_context_rows = FALSE; diff --git a/src/3rdparty/libjpeg/src/jdmerge.h b/src/3rdparty/libjpeg/src/jdmerge.h new file mode 100644 index 0000000000..b583396b10 --- /dev/null +++ b/src/3rdparty/libjpeg/src/jdmerge.h @@ -0,0 +1,47 @@ +/* + * jdmerge.h + * + * This file was part of the Independent JPEG Group's software: + * Copyright (C) 1994-1996, Thomas G. Lane. + * libjpeg-turbo Modifications: + * Copyright (C) 2020, D. R. Commander. + * For conditions of distribution and use, see the accompanying README.ijg + * file. + */ + +#define JPEG_INTERNALS +#include "jpeglib.h" + +#ifdef UPSAMPLE_MERGING_SUPPORTED + + +/* Private subobject */ + +typedef struct { + struct jpeg_upsampler pub; /* public fields */ + + /* Pointer to routine to do actual upsampling/conversion of one row group */ + void (*upmethod) (j_decompress_ptr cinfo, JSAMPIMAGE input_buf, + JDIMENSION in_row_group_ctr, JSAMPARRAY output_buf); + + /* Private state for YCC->RGB conversion */ + int *Cr_r_tab; /* => table for Cr to R conversion */ + int *Cb_b_tab; /* => table for Cb to B conversion */ + JLONG *Cr_g_tab; /* => table for Cr to G conversion */ + JLONG *Cb_g_tab; /* => table for Cb to G conversion */ + + /* For 2:1 vertical sampling, we produce two output rows at a time. + * We need a "spare" row buffer to hold the second output row if the + * application provides just a one-row buffer; we also use the spare + * to discard the dummy last row if the image height is odd. + */ + JSAMPROW spare_row; + boolean spare_full; /* T if spare buffer is occupied */ + + JDIMENSION out_row_width; /* samples per output row */ + JDIMENSION rows_to_go; /* counts rows remaining in image */ +} my_merged_upsampler; + +typedef my_merged_upsampler *my_merged_upsample_ptr; + +#endif /* UPSAMPLE_MERGING_SUPPORTED */ diff --git a/src/3rdparty/libjpeg/src/jdmrg565.c b/src/3rdparty/libjpeg/src/jdmrg565.c index 1b87e3718d..980a4e216e 100644 --- a/src/3rdparty/libjpeg/src/jdmrg565.c +++ b/src/3rdparty/libjpeg/src/jdmrg565.c @@ -5,7 +5,7 @@ * Copyright (C) 1994-1996, Thomas G. Lane. * libjpeg-turbo Modifications: * Copyright (C) 2013, Linaro Limited. - * Copyright (C) 2014-2015, 2018, D. R. Commander. + * Copyright (C) 2014-2015, 2018, 2020, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -19,7 +19,7 @@ h2v1_merged_upsample_565_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr, JSAMPARRAY output_buf) { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; register int y, cred, cgreen, cblue; int cb, cr; register JSAMPROW outptr; @@ -43,20 +43,20 @@ h2v1_merged_upsample_565_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, /* Loop for each pair of output pixels */ for (col = cinfo->output_width >> 1; col > 0; col--) { /* Do the chroma part of the calculation */ - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + cb = *inptr1++; + cr = *inptr2++; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; /* Fetch 2 Y values and emit 2 pixels */ - y = GETJSAMPLE(*inptr0++); + y = *inptr0++; r = range_limit[y + cred]; g = range_limit[y + cgreen]; b = range_limit[y + cblue]; rgb = PACK_SHORT_565(r, g, b); - y = GETJSAMPLE(*inptr0++); + y = *inptr0++; r = range_limit[y + cred]; g = range_limit[y + cgreen]; b = range_limit[y + cblue]; @@ -68,12 +68,12 @@ h2v1_merged_upsample_565_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, /* If image width is odd, do the last output column separately */ if (cinfo->output_width & 1) { - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); + cb = *inptr1; + cr = *inptr2; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; - y = GETJSAMPLE(*inptr0); + y = *inptr0; r = range_limit[y + cred]; g = range_limit[y + cgreen]; b = range_limit[y + cblue]; @@ -90,7 +90,7 @@ h2v1_merged_upsample_565D_internal(j_decompress_ptr cinfo, JDIMENSION in_row_group_ctr, JSAMPARRAY output_buf) { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; register int y, cred, cgreen, cblue; int cb, cr; register JSAMPROW outptr; @@ -115,21 +115,21 @@ h2v1_merged_upsample_565D_internal(j_decompress_ptr cinfo, /* Loop for each pair of output pixels */ for (col = cinfo->output_width >> 1; col > 0; col--) { /* Do the chroma part of the calculation */ - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + cb = *inptr1++; + cr = *inptr2++; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; /* Fetch 2 Y values and emit 2 pixels */ - y = GETJSAMPLE(*inptr0++); + y = *inptr0++; r = range_limit[DITHER_565_R(y + cred, d0)]; g = range_limit[DITHER_565_G(y + cgreen, d0)]; b = range_limit[DITHER_565_B(y + cblue, d0)]; d0 = DITHER_ROTATE(d0); rgb = PACK_SHORT_565(r, g, b); - y = GETJSAMPLE(*inptr0++); + y = *inptr0++; r = range_limit[DITHER_565_R(y + cred, d0)]; g = range_limit[DITHER_565_G(y + cgreen, d0)]; b = range_limit[DITHER_565_B(y + cblue, d0)]; @@ -142,12 +142,12 @@ h2v1_merged_upsample_565D_internal(j_decompress_ptr cinfo, /* If image width is odd, do the last output column separately */ if (cinfo->output_width & 1) { - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); + cb = *inptr1; + cr = *inptr2; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; - y = GETJSAMPLE(*inptr0); + y = *inptr0; r = range_limit[DITHER_565_R(y + cred, d0)]; g = range_limit[DITHER_565_G(y + cgreen, d0)]; b = range_limit[DITHER_565_B(y + cblue, d0)]; @@ -163,7 +163,7 @@ h2v2_merged_upsample_565_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr, JSAMPARRAY output_buf) { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; register int y, cred, cgreen, cblue; int cb, cr; register JSAMPROW outptr0, outptr1; @@ -189,20 +189,20 @@ h2v2_merged_upsample_565_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, /* Loop for each group of output pixels */ for (col = cinfo->output_width >> 1; col > 0; col--) { /* Do the chroma part of the calculation */ - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + cb = *inptr1++; + cr = *inptr2++; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; /* Fetch 4 Y values and emit 4 pixels */ - y = GETJSAMPLE(*inptr00++); + y = *inptr00++; r = range_limit[y + cred]; g = range_limit[y + cgreen]; b = range_limit[y + cblue]; rgb = PACK_SHORT_565(r, g, b); - y = GETJSAMPLE(*inptr00++); + y = *inptr00++; r = range_limit[y + cred]; g = range_limit[y + cgreen]; b = range_limit[y + cblue]; @@ -211,13 +211,13 @@ h2v2_merged_upsample_565_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, WRITE_TWO_PIXELS(outptr0, rgb); outptr0 += 4; - y = GETJSAMPLE(*inptr01++); + y = *inptr01++; r = range_limit[y + cred]; g = range_limit[y + cgreen]; b = range_limit[y + cblue]; rgb = PACK_SHORT_565(r, g, b); - y = GETJSAMPLE(*inptr01++); + y = *inptr01++; r = range_limit[y + cred]; g = range_limit[y + cgreen]; b = range_limit[y + cblue]; @@ -229,20 +229,20 @@ h2v2_merged_upsample_565_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, /* If image width is odd, do the last output column separately */ if (cinfo->output_width & 1) { - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); + cb = *inptr1; + cr = *inptr2; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; - y = GETJSAMPLE(*inptr00); + y = *inptr00; r = range_limit[y + cred]; g = range_limit[y + cgreen]; b = range_limit[y + cblue]; rgb = PACK_SHORT_565(r, g, b); *(INT16 *)outptr0 = (INT16)rgb; - y = GETJSAMPLE(*inptr01); + y = *inptr01; r = range_limit[y + cred]; g = range_limit[y + cgreen]; b = range_limit[y + cblue]; @@ -259,7 +259,7 @@ h2v2_merged_upsample_565D_internal(j_decompress_ptr cinfo, JDIMENSION in_row_group_ctr, JSAMPARRAY output_buf) { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; register int y, cred, cgreen, cblue; int cb, cr; register JSAMPROW outptr0, outptr1; @@ -287,21 +287,21 @@ h2v2_merged_upsample_565D_internal(j_decompress_ptr cinfo, /* Loop for each group of output pixels */ for (col = cinfo->output_width >> 1; col > 0; col--) { /* Do the chroma part of the calculation */ - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + cb = *inptr1++; + cr = *inptr2++; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; /* Fetch 4 Y values and emit 4 pixels */ - y = GETJSAMPLE(*inptr00++); + y = *inptr00++; r = range_limit[DITHER_565_R(y + cred, d0)]; g = range_limit[DITHER_565_G(y + cgreen, d0)]; b = range_limit[DITHER_565_B(y + cblue, d0)]; d0 = DITHER_ROTATE(d0); rgb = PACK_SHORT_565(r, g, b); - y = GETJSAMPLE(*inptr00++); + y = *inptr00++; r = range_limit[DITHER_565_R(y + cred, d0)]; g = range_limit[DITHER_565_G(y + cgreen, d0)]; b = range_limit[DITHER_565_B(y + cblue, d0)]; @@ -311,14 +311,14 @@ h2v2_merged_upsample_565D_internal(j_decompress_ptr cinfo, WRITE_TWO_PIXELS(outptr0, rgb); outptr0 += 4; - y = GETJSAMPLE(*inptr01++); + y = *inptr01++; r = range_limit[DITHER_565_R(y + cred, d1)]; g = range_limit[DITHER_565_G(y + cgreen, d1)]; b = range_limit[DITHER_565_B(y + cblue, d1)]; d1 = DITHER_ROTATE(d1); rgb = PACK_SHORT_565(r, g, b); - y = GETJSAMPLE(*inptr01++); + y = *inptr01++; r = range_limit[DITHER_565_R(y + cred, d1)]; g = range_limit[DITHER_565_G(y + cgreen, d1)]; b = range_limit[DITHER_565_B(y + cblue, d1)]; @@ -331,20 +331,20 @@ h2v2_merged_upsample_565D_internal(j_decompress_ptr cinfo, /* If image width is odd, do the last output column separately */ if (cinfo->output_width & 1) { - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); + cb = *inptr1; + cr = *inptr2; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; - y = GETJSAMPLE(*inptr00); + y = *inptr00; r = range_limit[DITHER_565_R(y + cred, d0)]; g = range_limit[DITHER_565_G(y + cgreen, d0)]; b = range_limit[DITHER_565_B(y + cblue, d0)]; rgb = PACK_SHORT_565(r, g, b); *(INT16 *)outptr0 = (INT16)rgb; - y = GETJSAMPLE(*inptr01); + y = *inptr01; r = range_limit[DITHER_565_R(y + cred, d1)]; g = range_limit[DITHER_565_G(y + cgreen, d1)]; b = range_limit[DITHER_565_B(y + cblue, d1)]; diff --git a/src/3rdparty/libjpeg/src/jdmrgext.c b/src/3rdparty/libjpeg/src/jdmrgext.c index b1c27df56a..038abc75d7 100644 --- a/src/3rdparty/libjpeg/src/jdmrgext.c +++ b/src/3rdparty/libjpeg/src/jdmrgext.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1994-1996, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2011, 2015, D. R. Commander. + * Copyright (C) 2011, 2015, 2020, 2023, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -25,7 +25,7 @@ h2v1_merged_upsample_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr, JSAMPARRAY output_buf) { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; register int y, cred, cgreen, cblue; int cb, cr; register JSAMPROW outptr; @@ -46,42 +46,42 @@ h2v1_merged_upsample_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, /* Loop for each pair of output pixels */ for (col = cinfo->output_width >> 1; col > 0; col--) { /* Do the chroma part of the calculation */ - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + cb = *inptr1++; + cr = *inptr2++; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; /* Fetch 2 Y values and emit 2 pixels */ - y = GETJSAMPLE(*inptr0++); + y = *inptr0++; outptr[RGB_RED] = range_limit[y + cred]; outptr[RGB_GREEN] = range_limit[y + cgreen]; outptr[RGB_BLUE] = range_limit[y + cblue]; #ifdef RGB_ALPHA - outptr[RGB_ALPHA] = 0xFF; + outptr[RGB_ALPHA] = MAXJSAMPLE; #endif outptr += RGB_PIXELSIZE; - y = GETJSAMPLE(*inptr0++); + y = *inptr0++; outptr[RGB_RED] = range_limit[y + cred]; outptr[RGB_GREEN] = range_limit[y + cgreen]; outptr[RGB_BLUE] = range_limit[y + cblue]; #ifdef RGB_ALPHA - outptr[RGB_ALPHA] = 0xFF; + outptr[RGB_ALPHA] = MAXJSAMPLE; #endif outptr += RGB_PIXELSIZE; } /* If image width is odd, do the last output column separately */ if (cinfo->output_width & 1) { - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); + cb = *inptr1; + cr = *inptr2; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; - y = GETJSAMPLE(*inptr0); + y = *inptr0; outptr[RGB_RED] = range_limit[y + cred]; outptr[RGB_GREEN] = range_limit[y + cgreen]; outptr[RGB_BLUE] = range_limit[y + cblue]; #ifdef RGB_ALPHA - outptr[RGB_ALPHA] = 0xFF; + outptr[RGB_ALPHA] = MAXJSAMPLE; #endif } } @@ -97,7 +97,7 @@ h2v2_merged_upsample_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr, JSAMPARRAY output_buf) { - my_upsample_ptr upsample = (my_upsample_ptr)cinfo->upsample; + my_merged_upsample_ptr upsample = (my_merged_upsample_ptr)cinfo->upsample; register int y, cred, cgreen, cblue; int cb, cr; register JSAMPROW outptr0, outptr1; @@ -120,65 +120,65 @@ h2v2_merged_upsample_internal(j_decompress_ptr cinfo, JSAMPIMAGE input_buf, /* Loop for each group of output pixels */ for (col = cinfo->output_width >> 1; col > 0; col--) { /* Do the chroma part of the calculation */ - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); + cb = *inptr1++; + cr = *inptr2++; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; /* Fetch 4 Y values and emit 4 pixels */ - y = GETJSAMPLE(*inptr00++); + y = *inptr00++; outptr0[RGB_RED] = range_limit[y + cred]; outptr0[RGB_GREEN] = range_limit[y + cgreen]; outptr0[RGB_BLUE] = range_limit[y + cblue]; #ifdef RGB_ALPHA - outptr0[RGB_ALPHA] = 0xFF; + outptr0[RGB_ALPHA] = MAXJSAMPLE; #endif outptr0 += RGB_PIXELSIZE; - y = GETJSAMPLE(*inptr00++); + y = *inptr00++; outptr0[RGB_RED] = range_limit[y + cred]; outptr0[RGB_GREEN] = range_limit[y + cgreen]; outptr0[RGB_BLUE] = range_limit[y + cblue]; #ifdef RGB_ALPHA - outptr0[RGB_ALPHA] = 0xFF; + outptr0[RGB_ALPHA] = MAXJSAMPLE; #endif outptr0 += RGB_PIXELSIZE; - y = GETJSAMPLE(*inptr01++); + y = *inptr01++; outptr1[RGB_RED] = range_limit[y + cred]; outptr1[RGB_GREEN] = range_limit[y + cgreen]; outptr1[RGB_BLUE] = range_limit[y + cblue]; #ifdef RGB_ALPHA - outptr1[RGB_ALPHA] = 0xFF; + outptr1[RGB_ALPHA] = MAXJSAMPLE; #endif outptr1 += RGB_PIXELSIZE; - y = GETJSAMPLE(*inptr01++); + y = *inptr01++; outptr1[RGB_RED] = range_limit[y + cred]; outptr1[RGB_GREEN] = range_limit[y + cgreen]; outptr1[RGB_BLUE] = range_limit[y + cblue]; #ifdef RGB_ALPHA - outptr1[RGB_ALPHA] = 0xFF; + outptr1[RGB_ALPHA] = MAXJSAMPLE; #endif outptr1 += RGB_PIXELSIZE; } /* If image width is odd, do the last output column separately */ if (cinfo->output_width & 1) { - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); + cb = *inptr1; + cr = *inptr2; cred = Crrtab[cr]; cgreen = (int)RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); cblue = Cbbtab[cb]; - y = GETJSAMPLE(*inptr00); + y = *inptr00; outptr0[RGB_RED] = range_limit[y + cred]; outptr0[RGB_GREEN] = range_limit[y + cgreen]; outptr0[RGB_BLUE] = range_limit[y + cblue]; #ifdef RGB_ALPHA - outptr0[RGB_ALPHA] = 0xFF; + outptr0[RGB_ALPHA] = MAXJSAMPLE; #endif - y = GETJSAMPLE(*inptr01); + y = *inptr01; outptr1[RGB_RED] = range_limit[y + cred]; outptr1[RGB_GREEN] = range_limit[y + cgreen]; outptr1[RGB_BLUE] = range_limit[y + cblue]; #ifdef RGB_ALPHA - outptr1[RGB_ALPHA] = 0xFF; + outptr1[RGB_ALPHA] = MAXJSAMPLE; #endif } } diff --git a/src/3rdparty/libjpeg/src/jdphuff.c b/src/3rdparty/libjpeg/src/jdphuff.c index 9e82636bbd..9680ebcbd0 100644 --- a/src/3rdparty/libjpeg/src/jdphuff.c +++ b/src/3rdparty/libjpeg/src/jdphuff.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1995-1997, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2015-2016, 2018, D. R. Commander. + * Copyright (C) 2015-2016, 2018-2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -41,25 +41,6 @@ typedef struct { int last_dc_val[MAX_COMPS_IN_SCAN]; /* last DC coef for each component */ } savable_state; -/* This macro is to work around compilers with missing or broken - * structure assignment. You'll need to fix this code if you have - * such a compiler and you change MAX_COMPS_IN_SCAN. - */ - -#ifndef NO_STRUCT_ASSIGN -#define ASSIGN_STATE(dest, src) ((dest) = (src)) -#else -#if MAX_COMPS_IN_SCAN == 4 -#define ASSIGN_STATE(dest, src) \ - ((dest).EOBRUN = (src).EOBRUN, \ - (dest).last_dc_val[0] = (src).last_dc_val[0], \ - (dest).last_dc_val[1] = (src).last_dc_val[1], \ - (dest).last_dc_val[2] = (src).last_dc_val[2], \ - (dest).last_dc_val[3] = (src).last_dc_val[3]) -#endif -#endif - - typedef struct { struct jpeg_entropy_decoder pub; /* public fields */ @@ -102,7 +83,7 @@ start_pass_phuff_decoder(j_decompress_ptr cinfo) boolean is_DC_band, bad; int ci, coefi, tbl; d_derived_tbl **pdtbl; - int *coef_bit_ptr; + int *coef_bit_ptr, *prev_coef_bit_ptr; jpeg_component_info *compptr; is_DC_band = (cinfo->Ss == 0); @@ -143,8 +124,15 @@ start_pass_phuff_decoder(j_decompress_ptr cinfo) for (ci = 0; ci < cinfo->comps_in_scan; ci++) { int cindex = cinfo->cur_comp_info[ci]->component_index; coef_bit_ptr = &cinfo->coef_bits[cindex][0]; + prev_coef_bit_ptr = &cinfo->coef_bits[cindex + cinfo->num_components][0]; if (!is_DC_band && coef_bit_ptr[0] < 0) /* AC without prior DC scan */ WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, 0); + for (coefi = MIN(cinfo->Ss, 1); coefi <= MAX(cinfo->Se, 9); coefi++) { + if (cinfo->input_scan_number > 1) + prev_coef_bit_ptr[coefi] = coef_bit_ptr[coefi]; + else + prev_coef_bit_ptr[coefi] = 0; + } for (coefi = cinfo->Ss; coefi <= cinfo->Se; coefi++) { int expected = (coef_bit_ptr[coefi] < 0) ? 0 : coef_bit_ptr[coefi]; if (cinfo->Ah != expected) @@ -323,7 +311,7 @@ decode_mcu_DC_first(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) /* Load up working state */ BITREAD_LOAD_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(state, entropy->saved); + state = entropy->saved; /* Outer loop handles each block in the MCU */ @@ -356,11 +344,12 @@ decode_mcu_DC_first(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) /* Completed MCU, so update state */ BITREAD_SAVE_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(entropy->saved, state); + entropy->saved = state; } /* Account for restart interval (no-op if not using restarts) */ - entropy->restarts_to_go--; + if (cinfo->restart_interval) + entropy->restarts_to_go--; return TRUE; } @@ -444,7 +433,8 @@ decode_mcu_AC_first(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) } /* Account for restart interval (no-op if not using restarts) */ - entropy->restarts_to_go--; + if (cinfo->restart_interval) + entropy->restarts_to_go--; return TRUE; } @@ -495,7 +485,8 @@ decode_mcu_DC_refine(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) BITREAD_SAVE_STATE(cinfo, entropy->bitstate); /* Account for restart interval (no-op if not using restarts) */ - entropy->restarts_to_go--; + if (cinfo->restart_interval) + entropy->restarts_to_go--; return TRUE; } @@ -587,9 +578,9 @@ decode_mcu_AC_refine(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) if (GET_BITS(1)) { if ((*thiscoef & p1) == 0) { /* do nothing if already set it */ if (*thiscoef >= 0) - *thiscoef += p1; + *thiscoef += (JCOEF)p1; else - *thiscoef += m1; + *thiscoef += (JCOEF)m1; } } } else { @@ -621,9 +612,9 @@ decode_mcu_AC_refine(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) if (GET_BITS(1)) { if ((*thiscoef & p1) == 0) { /* do nothing if already changed it */ if (*thiscoef >= 0) - *thiscoef += p1; + *thiscoef += (JCOEF)p1; else - *thiscoef += m1; + *thiscoef += (JCOEF)m1; } } } @@ -638,7 +629,8 @@ decode_mcu_AC_refine(j_decompress_ptr cinfo, JBLOCKROW *MCU_data) } /* Account for restart interval (no-op if not using restarts) */ - entropy->restarts_to_go--; + if (cinfo->restart_interval) + entropy->restarts_to_go--; return TRUE; @@ -676,7 +668,7 @@ jinit_phuff_decoder(j_decompress_ptr cinfo) /* Create progression status table */ cinfo->coef_bits = (int (*)[DCTSIZE2]) (*cinfo->mem->alloc_small) ((j_common_ptr)cinfo, JPOOL_IMAGE, - cinfo->num_components * DCTSIZE2 * + cinfo->num_components * 2 * DCTSIZE2 * sizeof(int)); coef_bit_ptr = &cinfo->coef_bits[0][0]; for (ci = 0; ci < cinfo->num_components; ci++) diff --git a/src/3rdparty/libjpeg/src/jdsample.c b/src/3rdparty/libjpeg/src/jdsample.c index 50a68b3013..eaad72a030 100644 --- a/src/3rdparty/libjpeg/src/jdsample.c +++ b/src/3rdparty/libjpeg/src/jdsample.c @@ -8,7 +8,7 @@ * Copyright (C) 2010, 2015-2016, D. R. Commander. * Copyright (C) 2014, MIPS Technologies, Inc., California. * Copyright (C) 2015, Google, Inc. - * Copyright (C) 2019, Arm Limited. + * Copyright (C) 2019-2020, Arm Limited. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -177,7 +177,7 @@ int_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, outptr = output_data[outrow]; outend = outptr + cinfo->output_width; while (outptr < outend) { - invalue = *inptr++; /* don't need GETJSAMPLE() here */ + invalue = *inptr++; for (h = h_expand; h > 0; h--) { *outptr++ = invalue; } @@ -213,7 +213,7 @@ h2v1_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, outptr = output_data[inrow]; outend = outptr + cinfo->output_width; while (outptr < outend) { - invalue = *inptr++; /* don't need GETJSAMPLE() here */ + invalue = *inptr++; *outptr++ = invalue; *outptr++ = invalue; } @@ -242,7 +242,7 @@ h2v2_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, outptr = output_data[outrow]; outend = outptr + cinfo->output_width; while (outptr < outend) { - invalue = *inptr++; /* don't need GETJSAMPLE() here */ + invalue = *inptr++; *outptr++ = invalue; *outptr++ = invalue; } @@ -283,20 +283,20 @@ h2v1_fancy_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, inptr = input_data[inrow]; outptr = output_data[inrow]; /* Special case for first column */ - invalue = GETJSAMPLE(*inptr++); + invalue = *inptr++; *outptr++ = (JSAMPLE)invalue; - *outptr++ = (JSAMPLE)((invalue * 3 + GETJSAMPLE(*inptr) + 2) >> 2); + *outptr++ = (JSAMPLE)((invalue * 3 + inptr[0] + 2) >> 2); for (colctr = compptr->downsampled_width - 2; colctr > 0; colctr--) { /* General case: 3/4 * nearer pixel + 1/4 * further pixel */ - invalue = GETJSAMPLE(*inptr++) * 3; - *outptr++ = (JSAMPLE)((invalue + GETJSAMPLE(inptr[-2]) + 1) >> 2); - *outptr++ = (JSAMPLE)((invalue + GETJSAMPLE(*inptr) + 2) >> 2); + invalue = (*inptr++) * 3; + *outptr++ = (JSAMPLE)((invalue + inptr[-2] + 1) >> 2); + *outptr++ = (JSAMPLE)((invalue + inptr[0] + 2) >> 2); } /* Special case for last column */ - invalue = GETJSAMPLE(*inptr); - *outptr++ = (JSAMPLE)((invalue * 3 + GETJSAMPLE(inptr[-1]) + 1) >> 2); + invalue = *inptr; + *outptr++ = (JSAMPLE)((invalue * 3 + inptr[-1] + 1) >> 2); *outptr++ = (JSAMPLE)invalue; } } @@ -338,7 +338,7 @@ h1v2_fancy_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, outptr = output_data[outrow++]; for (colctr = 0; colctr < compptr->downsampled_width; colctr++) { - thiscolsum = GETJSAMPLE(*inptr0++) * 3 + GETJSAMPLE(*inptr1++); + thiscolsum = (*inptr0++) * 3 + (*inptr1++); *outptr++ = (JSAMPLE)((thiscolsum + bias) >> 2); } } @@ -381,8 +381,8 @@ h2v2_fancy_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, outptr = output_data[outrow++]; /* Special case for first column */ - thiscolsum = GETJSAMPLE(*inptr0++) * 3 + GETJSAMPLE(*inptr1++); - nextcolsum = GETJSAMPLE(*inptr0++) * 3 + GETJSAMPLE(*inptr1++); + thiscolsum = (*inptr0++) * 3 + (*inptr1++); + nextcolsum = (*inptr0++) * 3 + (*inptr1++); *outptr++ = (JSAMPLE)((thiscolsum * 4 + 8) >> 4); *outptr++ = (JSAMPLE)((thiscolsum * 3 + nextcolsum + 7) >> 4); lastcolsum = thiscolsum; thiscolsum = nextcolsum; @@ -390,7 +390,7 @@ h2v2_fancy_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, for (colctr = compptr->downsampled_width - 2; colctr > 0; colctr--) { /* General case: 3/4 * nearer pixel + 1/4 * further pixel in each */ /* dimension, thus 9/16, 3/16, 3/16, 1/16 overall */ - nextcolsum = GETJSAMPLE(*inptr0++) * 3 + GETJSAMPLE(*inptr1++); + nextcolsum = (*inptr0++) * 3 + (*inptr1++); *outptr++ = (JSAMPLE)((thiscolsum * 3 + lastcolsum + 8) >> 4); *outptr++ = (JSAMPLE)((thiscolsum * 3 + nextcolsum + 7) >> 4); lastcolsum = thiscolsum; thiscolsum = nextcolsum; @@ -477,7 +477,13 @@ jinit_upsampler(j_decompress_ptr cinfo) } else if (h_in_group == h_out_group && v_in_group * 2 == v_out_group && do_fancy) { /* Non-fancy upsampling is handled by the generic method */ - upsample->methods[ci] = h1v2_fancy_upsample; +#if defined(__arm__) || defined(__aarch64__) || \ + defined(_M_ARM) || defined(_M_ARM64) + if (jsimd_can_h1v2_fancy_upsample()) + upsample->methods[ci] = jsimd_h1v2_fancy_upsample; + else +#endif + upsample->methods[ci] = h1v2_fancy_upsample; upsample->pub.need_context_rows = TRUE; } else if (h_in_group * 2 == h_out_group && v_in_group * 2 == v_out_group) { diff --git a/src/3rdparty/libjpeg/src/jdtrans.c b/src/3rdparty/libjpeg/src/jdtrans.c index 56713efe64..d7ec4b83b3 100644 --- a/src/3rdparty/libjpeg/src/jdtrans.c +++ b/src/3rdparty/libjpeg/src/jdtrans.c @@ -3,8 +3,8 @@ * * This file was part of the Independent JPEG Group's software: * Copyright (C) 1995-1997, Thomas G. Lane. - * It was modified by The libjpeg-turbo Project to include only code relevant - * to libjpeg-turbo. + * libjpeg-turbo Modifications: + * Copyright (C) 2020, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -16,6 +16,7 @@ #define JPEG_INTERNALS #include "jinclude.h" #include "jpeglib.h" +#include "jpegcomp.h" /* Forward declarations */ diff --git a/src/3rdparty/libjpeg/src/jerror.c b/src/3rdparty/libjpeg/src/jerror.c index 936c4f5d80..d0ab5b88b0 100644 --- a/src/3rdparty/libjpeg/src/jerror.c +++ b/src/3rdparty/libjpeg/src/jerror.c @@ -3,8 +3,8 @@ * * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1998, Thomas G. Lane. - * It was modified by The libjpeg-turbo Project to include only code relevant - * to libjpeg-turbo. + * libjpeg-turbo Modifications: + * Copyright (C) 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -189,13 +189,13 @@ format_message(j_common_ptr cinfo, char *buffer) /* Format the message into the passed buffer */ if (isstring) - sprintf(buffer, msgtext, err->msg_parm.s); + SNPRINTF(buffer, JMSG_LENGTH_MAX, msgtext, err->msg_parm.s); else - sprintf(buffer, msgtext, - err->msg_parm.i[0], err->msg_parm.i[1], - err->msg_parm.i[2], err->msg_parm.i[3], - err->msg_parm.i[4], err->msg_parm.i[5], - err->msg_parm.i[6], err->msg_parm.i[7]); + SNPRINTF(buffer, JMSG_LENGTH_MAX, msgtext, + err->msg_parm.i[0], err->msg_parm.i[1], + err->msg_parm.i[2], err->msg_parm.i[3], + err->msg_parm.i[4], err->msg_parm.i[5], + err->msg_parm.i[6], err->msg_parm.i[7]); } diff --git a/src/3rdparty/libjpeg/src/jerror.h b/src/3rdparty/libjpeg/src/jerror.h index 933a3690fd..eb44a1140a 100644 --- a/src/3rdparty/libjpeg/src/jerror.h +++ b/src/3rdparty/libjpeg/src/jerror.h @@ -5,7 +5,7 @@ * Copyright (C) 1994-1997, Thomas G. Lane. * Modified 1997-2009 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2014, 2017, D. R. Commander. + * Copyright (C) 2014, 2017, 2021-2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -103,7 +103,7 @@ JMESSAGE(JERR_MISMATCHED_QUANT_TABLE, "Cannot transcode due to multiple use of quantization table %d") JMESSAGE(JERR_MISSING_DATA, "Scan script does not transmit all data") JMESSAGE(JERR_MODE_CHANGE, "Invalid color quantization mode change") -JMESSAGE(JERR_NOTIMPL, "Not implemented yet") +JMESSAGE(JERR_NOTIMPL, "Requested features are incompatible") JMESSAGE(JERR_NOT_COMPILED, "Requested feature was omitted at compile time") #if JPEG_LIB_VERSION >= 70 JMESSAGE(JERR_NO_ARITH_TABLE, "Arithmetic table 0x%02x was not defined") @@ -207,6 +207,10 @@ JMESSAGE(JWRN_ARITH_BAD_CODE, "Corrupt JPEG data: bad arithmetic code") #endif #endif JMESSAGE(JWRN_BOGUS_ICC, "Corrupt JPEG data: bad ICC marker") +#if JPEG_LIB_VERSION < 70 +JMESSAGE(JERR_BAD_DROP_SAMPLING, + "Component index %d: mismatching sampling ratio %d:%d, %d:%d, %c") +#endif #ifdef JMAKE_ENUM_LIST @@ -252,9 +256,19 @@ JMESSAGE(JWRN_BOGUS_ICC, "Corrupt JPEG data: bad ICC marker") (cinfo)->err->msg_parm.i[2] = (p3), \ (cinfo)->err->msg_parm.i[3] = (p4), \ (*(cinfo)->err->error_exit) ((j_common_ptr)(cinfo))) +#define ERREXIT6(cinfo, code, p1, p2, p3, p4, p5, p6) \ + ((cinfo)->err->msg_code = (code), \ + (cinfo)->err->msg_parm.i[0] = (p1), \ + (cinfo)->err->msg_parm.i[1] = (p2), \ + (cinfo)->err->msg_parm.i[2] = (p3), \ + (cinfo)->err->msg_parm.i[3] = (p4), \ + (cinfo)->err->msg_parm.i[4] = (p5), \ + (cinfo)->err->msg_parm.i[5] = (p6), \ + (*(cinfo)->err->error_exit) ((j_common_ptr)(cinfo))) #define ERREXITS(cinfo, code, str) \ ((cinfo)->err->msg_code = (code), \ strncpy((cinfo)->err->msg_parm.s, (str), JMSG_STR_PARM_MAX), \ + (cinfo)->err->msg_parm.s[JMSG_STR_PARM_MAX - 1] = '\0', \ (*(cinfo)->err->error_exit) ((j_common_ptr)(cinfo))) #define MAKESTMT(stuff) do { stuff } while (0) @@ -311,6 +325,7 @@ JMESSAGE(JWRN_BOGUS_ICC, "Corrupt JPEG data: bad ICC marker") #define TRACEMSS(cinfo, lvl, code, str) \ ((cinfo)->err->msg_code = (code), \ strncpy((cinfo)->err->msg_parm.s, (str), JMSG_STR_PARM_MAX), \ + (cinfo)->err->msg_parm.s[JMSG_STR_PARM_MAX - 1] = '\0', \ (*(cinfo)->err->emit_message) ((j_common_ptr)(cinfo), (lvl))) #endif /* JERROR_H */ diff --git a/src/3rdparty/libjpeg/src/jfdctint.c b/src/3rdparty/libjpeg/src/jfdctint.c index b47c3061ac..c95a3a7fb8 100644 --- a/src/3rdparty/libjpeg/src/jfdctint.c +++ b/src/3rdparty/libjpeg/src/jfdctint.c @@ -4,11 +4,11 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1996, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2015, D. R. Commander. + * Copyright (C) 2015, 2020, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * - * This file contains a slow-but-accurate integer implementation of the + * This file contains a slower but more accurate integer implementation of the * forward DCT (Discrete Cosine Transform). * * A 2-D DCT can be done by 1-D DCT on each row followed by 1-D DCT diff --git a/src/3rdparty/libjpeg/src/jidctint.c b/src/3rdparty/libjpeg/src/jidctint.c index 98425d5fd0..bb08748019 100644 --- a/src/3rdparty/libjpeg/src/jidctint.c +++ b/src/3rdparty/libjpeg/src/jidctint.c @@ -3,13 +3,13 @@ * * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1998, Thomas G. Lane. - * Modification developed 2002-2009 by Guido Vollbeding. + * Modification developed 2002-2018 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2015, D. R. Commander. + * Copyright (C) 2015, 2020, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * - * This file contains a slow-but-accurate integer implementation of the + * This file contains a slower but more accurate integer implementation of the * inverse DCT (Discrete Cosine Transform). In the IJG code, this routine * must also perform dequantization of the input coefficients. * @@ -417,7 +417,7 @@ jpeg_idct_islow(j_decompress_ptr cinfo, jpeg_component_info *compptr, /* * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 7x7 output block. + * producing a reduced-size 7x7 output block. * * Optimized algorithm with 12 multiplications in the 1-D kernel. * cK represents sqrt(2) * cos(K*pi/14). @@ -1258,7 +1258,7 @@ jpeg_idct_10x10(j_decompress_ptr cinfo, jpeg_component_info *compptr, /* * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 11x11 output block. + * producing an 11x11 output block. * * Optimized algorithm with 24 multiplications in the 1-D kernel. * cK represents sqrt(2) * cos(K*pi/22). @@ -2398,7 +2398,7 @@ jpeg_idct_16x16(j_decompress_ptr cinfo, jpeg_component_info *compptr, tmp0 = DEQUANTIZE(inptr[DCTSIZE * 0], quantptr[DCTSIZE * 0]); tmp0 = LEFT_SHIFT(tmp0, CONST_BITS); /* Add fudge factor here for final descale. */ - tmp0 += 1 << (CONST_BITS - PASS1_BITS - 1); + tmp0 += ONE << (CONST_BITS - PASS1_BITS - 1); z1 = DEQUANTIZE(inptr[DCTSIZE * 4], quantptr[DCTSIZE * 4]); tmp1 = MULTIPLY(z1, FIX(1.306562965)); /* c4[16] = c2[8] */ diff --git a/src/3rdparty/libjpeg/src/jinclude.h b/src/3rdparty/libjpeg/src/jinclude.h index c1bcf7d9da..e8d983ac17 100644 --- a/src/3rdparty/libjpeg/src/jinclude.h +++ b/src/3rdparty/libjpeg/src/jinclude.h @@ -3,8 +3,8 @@ * * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1994, Thomas G. Lane. - * It was modified by The libjpeg-turbo Project to include only code relevant - * to libjpeg-turbo. + * libjpeg-turbo Modifications: + * Copyright (C) 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -17,72 +17,129 @@ * JPEG library. Most applications need only include jpeglib.h. */ +#ifndef __JINCLUDE_H__ +#define __JINCLUDE_H__ /* Include auto-config file to find out which system include files we need. */ #include "jconfig.h" /* auto configuration options */ +#include "jconfigint.h" #define JCONFIG_INCLUDED /* so that jpeglib.h doesn't do it again */ /* - * We need the NULL macro and size_t typedef. - * On an ANSI-conforming system it is sufficient to include <stddef.h>. - * Otherwise, we get them from <stdlib.h> or <stdio.h>; we may have to - * pull in <sys/types.h> as well. * Note that the core JPEG library does not require <stdio.h>; * only the default error handler and data source/destination modules do. * But we must pull it in because of the references to FILE in jpeglib.h. * You can remove those references if you want to compile without <stdio.h>. */ -#ifdef HAVE_STDDEF_H #include <stddef.h> -#endif - -#ifdef HAVE_STDLIB_H #include <stdlib.h> -#endif - -#ifdef NEED_SYS_TYPES_H -#include <sys/types.h> -#endif - #include <stdio.h> +#include <string.h> /* - * We need memory copying and zeroing functions, plus strncpy(). - * ANSI and System V implementations declare these in <string.h>. - * BSD doesn't have the mem() functions, but it does have bcopy()/bzero(). - * Some systems may declare memset and memcpy in <memory.h>. - * - * NOTE: we assume the size parameters to these functions are of type size_t. - * Change the casts in these macros if not! + * These macros/inline functions facilitate using Microsoft's "safe string" + * functions with Visual Studio builds without the need to scatter #ifdefs + * throughout the code base. */ -#ifdef NEED_BSD_STRINGS -#include <strings.h> -#define MEMZERO(target, size) \ - bzero((void *)(target), (size_t)(size)) -#define MEMCOPY(dest, src, size) \ - bcopy((const void *)(src), (void *)(dest), (size_t)(size)) +#ifdef _MSC_VER -#else /* not BSD, assume ANSI/SysV string lib */ +#define SNPRINTF(str, n, format, ...) \ + _snprintf_s(str, n, _TRUNCATE, format, ##__VA_ARGS__) -#include <string.h> -#define MEMZERO(target, size) \ - memset((void *)(target), 0, (size_t)(size)) -#define MEMCOPY(dest, src, size) \ - memcpy((void *)(dest), (const void *)(src), (size_t)(size)) +#else + +#define SNPRINTF snprintf #endif -/* - * The modules that use fread() and fwrite() always invoke them through - * these macros. On some systems you may need to twiddle the argument casts. - * CAUTION: argument order is different from underlying functions! + +#ifndef NO_GETENV + +#ifdef _MSC_VER + +static INLINE int GETENV_S(char *buffer, size_t buffer_size, const char *name) +{ + size_t required_size; + + return (int)getenv_s(&required_size, buffer, buffer_size, name); +} + +#else /* _MSC_VER */ + +#include <errno.h> + +/* This provides a similar interface to the Microsoft/C11 getenv_s() function, + * but other than parameter validation, it has no advantages over getenv(). */ -#define JFREAD(file, buf, sizeofbuf) \ - ((size_t)fread((void *)(buf), (size_t)1, (size_t)(sizeofbuf), (file))) -#define JFWRITE(file, buf, sizeofbuf) \ - ((size_t)fwrite((const void *)(buf), (size_t)1, (size_t)(sizeofbuf), (file))) +static INLINE int GETENV_S(char *buffer, size_t buffer_size, const char *name) +{ + char *env; + + if (!buffer) { + if (buffer_size == 0) + return 0; + else + return (errno = EINVAL); + } + if (buffer_size == 0) + return (errno = EINVAL); + if (!name) { + *buffer = 0; + return 0; + } + + env = getenv(name); + if (!env) + { + *buffer = 0; + return 0; + } + + if (strlen(env) + 1 > buffer_size) { + *buffer = 0; + return ERANGE; + } + + strncpy(buffer, env, buffer_size); + + return 0; +} + +#endif /* _MSC_VER */ + +#endif /* NO_GETENV */ + + +#ifndef NO_PUTENV + +#ifdef _WIN32 + +#define PUTENV_S(name, value) _putenv_s(name, value) + +#else + +/* This provides a similar interface to the Microsoft _putenv_s() function, but + * other than parameter validation, it has no advantages over setenv(). + */ + +static INLINE int PUTENV_S(const char *name, const char *value) +{ + if (!name || !value) + return (errno = EINVAL); + + setenv(name, value, 1); + + return errno; +} + +#endif /* _WIN32 */ + +#endif /* NO_PUTENV */ + + +#endif /* JINCLUDE_H */ diff --git a/src/3rdparty/libjpeg/src/jmemmgr.c b/src/3rdparty/libjpeg/src/jmemmgr.c index 508ca7429c..a40446f6ac 100644 --- a/src/3rdparty/libjpeg/src/jmemmgr.c +++ b/src/3rdparty/libjpeg/src/jmemmgr.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1997, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2016, D. R. Commander. + * Copyright (C) 2016, 2021-2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -37,12 +37,6 @@ #endif #include <limits.h> -#ifndef NO_GETENV -#ifndef HAVE_STDLIB_H /* <stdlib.h> should declare getenv() */ -extern char *getenv(const char *name); -#endif -#endif - LOCAL(size_t) round_up_pow2(size_t a, size_t b) @@ -74,10 +68,13 @@ round_up_pow2(size_t a, size_t b) * There isn't any really portable way to determine the worst-case alignment * requirement. This module assumes that the alignment requirement is * multiples of ALIGN_SIZE. - * By default, we define ALIGN_SIZE as sizeof(double). This is necessary on - * some workstations (where doubles really do need 8-byte alignment) and will - * work fine on nearly everything. If your machine has lesser alignment needs, - * you can save a few bytes by making ALIGN_SIZE smaller. + * By default, we define ALIGN_SIZE as the maximum of sizeof(double) and + * sizeof(void *). This is necessary on some workstations (where doubles + * really do need 8-byte alignment) and will work fine on nearly everything. + * We use the maximum of sizeof(double) and sizeof(void *) since sizeof(double) + * may be insufficient, for example, on CHERI-enabled platforms with 16-byte + * pointers and a 16-byte alignment requirement. If your machine has lesser + * alignment needs, you can save a few bytes by making ALIGN_SIZE smaller. * The only place I know of where this will NOT work is certain Macintosh * 680x0 compilers that define double as a 10-byte IEEE extended float. * Doing 10-byte alignment is counterproductive because longwords won't be @@ -87,7 +84,7 @@ round_up_pow2(size_t a, size_t b) #ifndef ALIGN_SIZE /* so can override from jconfig.h */ #ifndef WITH_SIMD -#define ALIGN_SIZE sizeof(double) +#define ALIGN_SIZE MAX(sizeof(void *), sizeof(double)) #else #define ALIGN_SIZE 32 /* Most of the SIMD instructions we support require 16-byte (128-bit) alignment, but AVX2 requires @@ -1032,7 +1029,7 @@ free_pool(j_common_ptr cinfo, int pool_id) large_pool_ptr next_lhdr_ptr = lhdr_ptr->next; space_freed = lhdr_ptr->bytes_used + lhdr_ptr->bytes_left + - sizeof(large_pool_hdr); + sizeof(large_pool_hdr) + ALIGN_SIZE - 1; jpeg_free_large(cinfo, (void *)lhdr_ptr, space_freed); mem->total_space_allocated -= space_freed; lhdr_ptr = next_lhdr_ptr; @@ -1045,7 +1042,7 @@ free_pool(j_common_ptr cinfo, int pool_id) while (shdr_ptr != NULL) { small_pool_ptr next_shdr_ptr = shdr_ptr->next; space_freed = shdr_ptr->bytes_used + shdr_ptr->bytes_left + - sizeof(small_pool_hdr); + sizeof(small_pool_hdr) + ALIGN_SIZE - 1; jpeg_free_small(cinfo, (void *)shdr_ptr, space_freed); mem->total_space_allocated -= space_freed; shdr_ptr = next_shdr_ptr; @@ -1162,12 +1159,16 @@ jinit_memory_mgr(j_common_ptr cinfo) */ #ifndef NO_GETENV { - char *memenv; + char memenv[30] = { 0 }; - if ((memenv = getenv("JPEGMEM")) != NULL) { + if (!GETENV_S(memenv, 30, "JPEGMEM") && strlen(memenv) > 0) { char ch = 'x'; +#ifdef _MSC_VER + if (sscanf_s(memenv, "%ld%c", &max_to_use, &ch, 1) > 0) { +#else if (sscanf(memenv, "%ld%c", &max_to_use, &ch) > 0) { +#endif if (ch == 'm' || ch == 'M') max_to_use *= 1000L; mem->pub.max_memory_to_use = max_to_use * 1000L; diff --git a/src/3rdparty/libjpeg/src/jmemnobs.c b/src/3rdparty/libjpeg/src/jmemnobs.c index 089be8f500..cd6571ba1c 100644 --- a/src/3rdparty/libjpeg/src/jmemnobs.c +++ b/src/3rdparty/libjpeg/src/jmemnobs.c @@ -22,11 +22,6 @@ #include "jpeglib.h" #include "jmemsys.h" /* import the system-dependent declarations */ -#ifndef HAVE_STDLIB_H /* <stdlib.h> should declare malloc(),free() */ -extern void *malloc(size_t size); -extern void free(void *ptr); -#endif - /* * Memory allocation and freeing are controlled by the regular library diff --git a/src/3rdparty/libjpeg/src/jmorecfg.h b/src/3rdparty/libjpeg/src/jmorecfg.h index d0b930079a..b33a991914 100644 --- a/src/3rdparty/libjpeg/src/jmorecfg.h +++ b/src/3rdparty/libjpeg/src/jmorecfg.h @@ -5,7 +5,7 @@ * Copyright (C) 1991-1997, Thomas G. Lane. * Modified 1997-2009 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2009, 2011, 2014-2015, 2018, D. R. Commander. + * Copyright (C) 2009, 2011, 2014-2015, 2018, 2020, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -43,25 +43,11 @@ #if BITS_IN_JSAMPLE == 8 /* JSAMPLE should be the smallest type that will hold the values 0..255. - * You can use a signed char by having GETJSAMPLE mask it with 0xFF. */ -#ifdef HAVE_UNSIGNED_CHAR - typedef unsigned char JSAMPLE; #define GETJSAMPLE(value) ((int)(value)) -#else /* not HAVE_UNSIGNED_CHAR */ - -typedef char JSAMPLE; -#ifdef __CHAR_UNSIGNED__ -#define GETJSAMPLE(value) ((int)(value)) -#else -#define GETJSAMPLE(value) ((int)(value) & 0xFF) -#endif /* __CHAR_UNSIGNED__ */ - -#endif /* HAVE_UNSIGNED_CHAR */ - #define MAXJSAMPLE 255 #define CENTERJSAMPLE 128 @@ -97,22 +83,9 @@ typedef short JCOEF; * managers, this is also the data type passed to fread/fwrite. */ -#ifdef HAVE_UNSIGNED_CHAR - typedef unsigned char JOCTET; #define GETJOCTET(value) (value) -#else /* not HAVE_UNSIGNED_CHAR */ - -typedef char JOCTET; -#ifdef __CHAR_UNSIGNED__ -#define GETJOCTET(value) (value) -#else -#define GETJOCTET(value) ((value) & 0xFF) -#endif /* __CHAR_UNSIGNED__ */ - -#endif /* HAVE_UNSIGNED_CHAR */ - /* These typedefs are used for various table entries and so forth. * They must be at least as wide as specified; but making them too big @@ -123,23 +96,11 @@ typedef char JOCTET; /* UINT8 must hold at least the values 0..255. */ -#ifdef HAVE_UNSIGNED_CHAR typedef unsigned char UINT8; -#else /* not HAVE_UNSIGNED_CHAR */ -#ifdef __CHAR_UNSIGNED__ -typedef char UINT8; -#else /* not __CHAR_UNSIGNED__ */ -typedef short UINT8; -#endif /* __CHAR_UNSIGNED__ */ -#endif /* HAVE_UNSIGNED_CHAR */ /* UINT16 must hold at least the values 0..65535. */ -#ifdef HAVE_UNSIGNED_SHORT typedef unsigned short UINT16; -#else /* not HAVE_UNSIGNED_SHORT */ -typedef unsigned int UINT16; -#endif /* HAVE_UNSIGNED_SHORT */ /* INT16 must hold at least the values -32768..32767. */ @@ -273,9 +234,9 @@ typedef int boolean; /* Capability options common to encoder and decoder: */ -#define DCT_ISLOW_SUPPORTED /* slow but accurate integer algorithm */ -#define DCT_IFAST_SUPPORTED /* faster, less accurate integer method */ -#define DCT_FLOAT_SUPPORTED /* floating-point: accurate, fast on fast HW */ +#define DCT_ISLOW_SUPPORTED /* accurate integer method */ +#define DCT_IFAST_SUPPORTED /* less accurate int method [legacy feature] */ +#define DCT_FLOAT_SUPPORTED /* floating-point method [legacy feature] */ /* Encoder capability options: */ diff --git a/src/3rdparty/libjpeg/src/jpegcomp.h b/src/3rdparty/libjpeg/src/jpegcomp.h index b32d544bf1..c4834ac0df 100644 --- a/src/3rdparty/libjpeg/src/jpegcomp.h +++ b/src/3rdparty/libjpeg/src/jpegcomp.h @@ -1,7 +1,7 @@ /* * jpegcomp.h * - * Copyright (C) 2010, D. R. Commander. + * Copyright (C) 2010, 2020, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -19,6 +19,7 @@ #define _min_DCT_v_scaled_size min_DCT_v_scaled_size #define _jpeg_width jpeg_width #define _jpeg_height jpeg_height +#define JERR_ARITH_NOTIMPL JERR_NOT_COMPILED #else #define _DCT_scaled_size DCT_scaled_size #define _DCT_h_scaled_size DCT_scaled_size diff --git a/src/3rdparty/libjpeg/src/jpegint.h b/src/3rdparty/libjpeg/src/jpegint.h index ad36ca8b56..6af9e2a179 100644 --- a/src/3rdparty/libjpeg/src/jpegint.h +++ b/src/3rdparty/libjpeg/src/jpegint.h @@ -5,8 +5,9 @@ * Copyright (C) 1991-1997, Thomas G. Lane. * Modified 1997-2009 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2015-2016, D. R. Commander. + * Copyright (C) 2015-2016, 2019, 2021, D. R. Commander. * Copyright (C) 2015, Google, Inc. + * Copyright (C) 2021, Alex Richardson. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -47,6 +48,18 @@ typedef enum { /* Operating modes for buffer controllers */ /* JLONG must hold at least signed 32-bit values. */ typedef long JLONG; +/* JUINTPTR must hold pointer values. */ +#ifdef __UINTPTR_TYPE__ +/* + * __UINTPTR_TYPE__ is GNU-specific and available in GCC 4.6+ and Clang 3.0+. + * Fortunately, that is sufficient to support the few architectures for which + * sizeof(void *) != sizeof(size_t). The only other options would require C99 + * or Clang-specific builtins. + */ +typedef __UINTPTR_TYPE__ JUINTPTR; +#else +typedef size_t JUINTPTR; +#endif /* * Left shift macro that handles a negative operand without causing any @@ -158,6 +171,9 @@ struct jpeg_decomp_master { JDIMENSION first_MCU_col[MAX_COMPONENTS]; JDIMENSION last_MCU_col[MAX_COMPONENTS]; boolean jinit_upsampler_no_alloc; + + /* Last iMCU row that was successfully decoded */ + JDIMENSION last_good_iMCU_row; }; /* Input control module */ @@ -357,12 +373,3 @@ extern const int jpeg_natural_order[]; /* zigzag coef order to natural order */ /* Arithmetic coding probability estimation tables in jaricom.c */ extern const JLONG jpeg_aritab[]; - -/* Suppress undefined-structure complaints if necessary. */ - -#ifdef INCOMPLETE_TYPES_BROKEN -#ifndef AM_MEMORY_MANAGER /* only jmemmgr.c defines these */ -struct jvirt_sarray_control { long dummy; }; -struct jvirt_barray_control { long dummy; }; -#endif -#endif /* INCOMPLETE_TYPES_BROKEN */ diff --git a/src/3rdparty/libjpeg/src/jpeglib.h b/src/3rdparty/libjpeg/src/jpeglib.h index 33f8ad2791..d7664f0630 100644 --- a/src/3rdparty/libjpeg/src/jpeglib.h +++ b/src/3rdparty/libjpeg/src/jpeglib.h @@ -5,7 +5,7 @@ * Copyright (C) 1991-1998, Thomas G. Lane. * Modified 2002-2009 by Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2009-2011, 2013-2014, 2016-2017, D. R. Commander. + * Copyright (C) 2009-2011, 2013-2014, 2016-2017, 2020, D. R. Commander. * Copyright (C) 2015, Google, Inc. * For conditions of distribution and use, see the accompanying README.ijg * file. @@ -244,9 +244,9 @@ typedef enum { /* DCT/IDCT algorithm options. */ typedef enum { - JDCT_ISLOW, /* slow but accurate integer algorithm */ - JDCT_IFAST, /* faster, less accurate integer method */ - JDCT_FLOAT /* floating-point: accurate, fast on fast HW */ + JDCT_ISLOW, /* accurate integer method */ + JDCT_IFAST, /* less accurate integer method [legacy feature] */ + JDCT_FLOAT /* floating-point method [legacy feature] */ } J_DCT_METHOD; #ifndef JDCT_DEFAULT /* may be overridden in jconfig.h */ diff --git a/src/3rdparty/libjpeg/src/jquant1.c b/src/3rdparty/libjpeg/src/jquant1.c index 40bbb28cc7..73b83e16e5 100644 --- a/src/3rdparty/libjpeg/src/jquant1.c +++ b/src/3rdparty/libjpeg/src/jquant1.c @@ -479,7 +479,7 @@ color_quantize(j_decompress_ptr cinfo, JSAMPARRAY input_buf, for (col = width; col > 0; col--) { pixcode = 0; for (ci = 0; ci < nc; ci++) { - pixcode += GETJSAMPLE(colorindex[ci][GETJSAMPLE(*ptrin++)]); + pixcode += colorindex[ci][*ptrin++]; } *ptrout++ = (JSAMPLE)pixcode; } @@ -506,9 +506,9 @@ color_quantize3(j_decompress_ptr cinfo, JSAMPARRAY input_buf, ptrin = input_buf[row]; ptrout = output_buf[row]; for (col = width; col > 0; col--) { - pixcode = GETJSAMPLE(colorindex0[GETJSAMPLE(*ptrin++)]); - pixcode += GETJSAMPLE(colorindex1[GETJSAMPLE(*ptrin++)]); - pixcode += GETJSAMPLE(colorindex2[GETJSAMPLE(*ptrin++)]); + pixcode = colorindex0[*ptrin++]; + pixcode += colorindex1[*ptrin++]; + pixcode += colorindex2[*ptrin++]; *ptrout++ = (JSAMPLE)pixcode; } } @@ -552,7 +552,7 @@ quantize_ord_dither(j_decompress_ptr cinfo, JSAMPARRAY input_buf, * required amount of padding. */ *output_ptr += - colorindex_ci[GETJSAMPLE(*input_ptr) + dither[col_index]]; + colorindex_ci[*input_ptr + dither[col_index]]; input_ptr += nc; output_ptr++; col_index = (col_index + 1) & ODITHER_MASK; @@ -595,12 +595,9 @@ quantize3_ord_dither(j_decompress_ptr cinfo, JSAMPARRAY input_buf, col_index = 0; for (col = width; col > 0; col--) { - pixcode = - GETJSAMPLE(colorindex0[GETJSAMPLE(*input_ptr++) + dither0[col_index]]); - pixcode += - GETJSAMPLE(colorindex1[GETJSAMPLE(*input_ptr++) + dither1[col_index]]); - pixcode += - GETJSAMPLE(colorindex2[GETJSAMPLE(*input_ptr++) + dither2[col_index]]); + pixcode = colorindex0[(*input_ptr++) + dither0[col_index]]; + pixcode += colorindex1[(*input_ptr++) + dither1[col_index]]; + pixcode += colorindex2[(*input_ptr++) + dither2[col_index]]; *output_ptr++ = (JSAMPLE)pixcode; col_index = (col_index + 1) & ODITHER_MASK; } @@ -677,15 +674,15 @@ quantize_fs_dither(j_decompress_ptr cinfo, JSAMPARRAY input_buf, * The maximum error is +- MAXJSAMPLE; this sets the required size * of the range_limit array. */ - cur += GETJSAMPLE(*input_ptr); - cur = GETJSAMPLE(range_limit[cur]); + cur += *input_ptr; + cur = range_limit[cur]; /* Select output value, accumulate into output code for this pixel */ - pixcode = GETJSAMPLE(colorindex_ci[cur]); + pixcode = colorindex_ci[cur]; *output_ptr += (JSAMPLE)pixcode; /* Compute actual representation error at this pixel */ /* Note: we can do this even though we don't have the final */ /* pixel code, because the colormap is orthogonal. */ - cur -= GETJSAMPLE(colormap_ci[pixcode]); + cur -= colormap_ci[pixcode]; /* Compute error fractions to be propagated to adjacent pixels. * Add these into the running sums, and simultaneously shift the * next-line error sums left by 1 column. diff --git a/src/3rdparty/libjpeg/src/jquant2.c b/src/3rdparty/libjpeg/src/jquant2.c index 0ce0ca5472..44efb18cad 100644 --- a/src/3rdparty/libjpeg/src/jquant2.c +++ b/src/3rdparty/libjpeg/src/jquant2.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1996, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2009, 2014-2015, D. R. Commander. + * Copyright (C) 2009, 2014-2015, 2020, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -215,9 +215,9 @@ prescan_quantize(j_decompress_ptr cinfo, JSAMPARRAY input_buf, ptr = input_buf[row]; for (col = width; col > 0; col--) { /* get pixel value and index into the histogram */ - histp = &histogram[GETJSAMPLE(ptr[0]) >> C0_SHIFT] - [GETJSAMPLE(ptr[1]) >> C1_SHIFT] - [GETJSAMPLE(ptr[2]) >> C2_SHIFT]; + histp = &histogram[ptr[0] >> C0_SHIFT] + [ptr[1] >> C1_SHIFT] + [ptr[2] >> C2_SHIFT]; /* increment, check for overflow and undo increment if so. */ if (++(*histp) <= 0) (*histp)--; @@ -665,7 +665,7 @@ find_nearby_colors(j_decompress_ptr cinfo, int minc0, int minc1, int minc2, for (i = 0; i < numcolors; i++) { /* We compute the squared-c0-distance term, then add in the other two. */ - x = GETJSAMPLE(cinfo->colormap[0][i]); + x = cinfo->colormap[0][i]; if (x < minc0) { tdist = (x - minc0) * C0_SCALE; min_dist = tdist * tdist; @@ -688,7 +688,7 @@ find_nearby_colors(j_decompress_ptr cinfo, int minc0, int minc1, int minc2, } } - x = GETJSAMPLE(cinfo->colormap[1][i]); + x = cinfo->colormap[1][i]; if (x < minc1) { tdist = (x - minc1) * C1_SCALE; min_dist += tdist * tdist; @@ -710,7 +710,7 @@ find_nearby_colors(j_decompress_ptr cinfo, int minc0, int minc1, int minc2, } } - x = GETJSAMPLE(cinfo->colormap[2][i]); + x = cinfo->colormap[2][i]; if (x < minc2) { tdist = (x - minc2) * C2_SCALE; min_dist += tdist * tdist; @@ -788,13 +788,13 @@ find_best_colors(j_decompress_ptr cinfo, int minc0, int minc1, int minc2, #define STEP_C2 ((1 << C2_SHIFT) * C2_SCALE) for (i = 0; i < numcolors; i++) { - icolor = GETJSAMPLE(colorlist[i]); + icolor = colorlist[i]; /* Compute (square of) distance from minc0/c1/c2 to this color */ - inc0 = (minc0 - GETJSAMPLE(cinfo->colormap[0][icolor])) * C0_SCALE; + inc0 = (minc0 - cinfo->colormap[0][icolor]) * C0_SCALE; dist0 = inc0 * inc0; - inc1 = (minc1 - GETJSAMPLE(cinfo->colormap[1][icolor])) * C1_SCALE; + inc1 = (minc1 - cinfo->colormap[1][icolor]) * C1_SCALE; dist0 += inc1 * inc1; - inc2 = (minc2 - GETJSAMPLE(cinfo->colormap[2][icolor])) * C2_SCALE; + inc2 = (minc2 - cinfo->colormap[2][icolor]) * C2_SCALE; dist0 += inc2 * inc2; /* Form the initial difference increments */ inc0 = inc0 * (2 * STEP_C0) + STEP_C0 * STEP_C0; @@ -879,7 +879,7 @@ fill_inverse_cmap(j_decompress_ptr cinfo, int c0, int c1, int c2) for (ic1 = 0; ic1 < BOX_C1_ELEMS; ic1++) { cachep = &histogram[c0 + ic0][c1 + ic1][c2]; for (ic2 = 0; ic2 < BOX_C2_ELEMS; ic2++) { - *cachep++ = (histcell)(GETJSAMPLE(*cptr++) + 1); + *cachep++ = (histcell)((*cptr++) + 1); } } } @@ -909,9 +909,9 @@ pass2_no_dither(j_decompress_ptr cinfo, JSAMPARRAY input_buf, outptr = output_buf[row]; for (col = width; col > 0; col--) { /* get pixel value and index into the cache */ - c0 = GETJSAMPLE(*inptr++) >> C0_SHIFT; - c1 = GETJSAMPLE(*inptr++) >> C1_SHIFT; - c2 = GETJSAMPLE(*inptr++) >> C2_SHIFT; + c0 = (*inptr++) >> C0_SHIFT; + c1 = (*inptr++) >> C1_SHIFT; + c2 = (*inptr++) >> C2_SHIFT; cachep = &histogram[c0][c1][c2]; /* If we have not seen this color before, find nearest colormap entry */ /* and update the cache */ @@ -996,12 +996,12 @@ pass2_fs_dither(j_decompress_ptr cinfo, JSAMPARRAY input_buf, * The maximum error is +- MAXJSAMPLE (or less with error limiting); * this sets the required size of the range_limit array. */ - cur0 += GETJSAMPLE(inptr[0]); - cur1 += GETJSAMPLE(inptr[1]); - cur2 += GETJSAMPLE(inptr[2]); - cur0 = GETJSAMPLE(range_limit[cur0]); - cur1 = GETJSAMPLE(range_limit[cur1]); - cur2 = GETJSAMPLE(range_limit[cur2]); + cur0 += inptr[0]; + cur1 += inptr[1]; + cur2 += inptr[2]; + cur0 = range_limit[cur0]; + cur1 = range_limit[cur1]; + cur2 = range_limit[cur2]; /* Index into the cache with adjusted pixel value */ cachep = &histogram[cur0 >> C0_SHIFT][cur1 >> C1_SHIFT][cur2 >> C2_SHIFT]; @@ -1015,9 +1015,9 @@ pass2_fs_dither(j_decompress_ptr cinfo, JSAMPARRAY input_buf, register int pixcode = *cachep - 1; *outptr = (JSAMPLE)pixcode; /* Compute representation error for this pixel */ - cur0 -= GETJSAMPLE(colormap0[pixcode]); - cur1 -= GETJSAMPLE(colormap1[pixcode]); - cur2 -= GETJSAMPLE(colormap2[pixcode]); + cur0 -= colormap0[pixcode]; + cur1 -= colormap1[pixcode]; + cur2 -= colormap2[pixcode]; } /* Compute error fractions to be propagated to adjacent pixels. * Add these into the running sums, and simultaneously shift the @@ -1145,7 +1145,7 @@ start_pass_2_quant(j_decompress_ptr cinfo, boolean is_pre_scan) int i; /* Only F-S dithering or no dithering is supported. */ - /* If user asks for ordered dither, give him F-S. */ + /* If user asks for ordered dither, give them F-S. */ if (cinfo->dither_mode != JDITHER_NONE) cinfo->dither_mode = JDITHER_FS; @@ -1263,7 +1263,7 @@ jinit_2pass_quantizer(j_decompress_ptr cinfo) cquantize->sv_colormap = NULL; /* Only F-S dithering or no dithering is supported. */ - /* If user asks for ordered dither, give him F-S. */ + /* If user asks for ordered dither, give them F-S. */ if (cinfo->dither_mode != JDITHER_NONE) cinfo->dither_mode = JDITHER_FS; diff --git a/src/3rdparty/libjpeg/src/jsimd.h b/src/3rdparty/libjpeg/src/jsimd.h index 51e2b8c89d..74d480aa2c 100644 --- a/src/3rdparty/libjpeg/src/jsimd.h +++ b/src/3rdparty/libjpeg/src/jsimd.h @@ -2,8 +2,9 @@ * jsimd.h * * Copyright 2009 Pierre Ossman <ossman@cendio.se> for Cendio AB - * Copyright (C) 2011, 2014, D. R. Commander. - * Copyright (C) 2015-2016, 2018, Matthieu Darbois. + * Copyright (C) 2011, 2014, 2022, D. R. Commander. + * Copyright (C) 2015-2016, 2018, 2022, Matthieu Darbois. + * Copyright (C) 2020, Arm Limited. * * Based on the x86 SIMD extension for IJG JPEG library, * Copyright (C) 1999-2006, MIYASAKA Masaru. @@ -75,6 +76,7 @@ EXTERN(void) jsimd_int_upsample(j_decompress_ptr cinfo, EXTERN(int) jsimd_can_h2v2_fancy_upsample(void); EXTERN(int) jsimd_can_h2v1_fancy_upsample(void); +EXTERN(int) jsimd_can_h1v2_fancy_upsample(void); EXTERN(void) jsimd_h2v2_fancy_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, @@ -84,6 +86,10 @@ EXTERN(void) jsimd_h2v1_fancy_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, JSAMPARRAY input_data, JSAMPARRAY *output_data_ptr); +EXTERN(void) jsimd_h1v2_fancy_upsample(j_decompress_ptr cinfo, + jpeg_component_info *compptr, + JSAMPARRAY input_data, + JSAMPARRAY *output_data_ptr); EXTERN(int) jsimd_can_h2v2_merged_upsample(void); EXTERN(int) jsimd_can_h2v1_merged_upsample(void); @@ -108,10 +114,10 @@ EXTERN(int) jsimd_can_encode_mcu_AC_first_prepare(void); EXTERN(void) jsimd_encode_mcu_AC_first_prepare (const JCOEF *block, const int *jpeg_natural_order_start, int Sl, int Al, - JCOEF *values, size_t *zerobits); + UJCOEF *values, size_t *zerobits); EXTERN(int) jsimd_can_encode_mcu_AC_refine_prepare(void); EXTERN(int) jsimd_encode_mcu_AC_refine_prepare (const JCOEF *block, const int *jpeg_natural_order_start, int Sl, int Al, - JCOEF *absvalues, size_t *bits); + UJCOEF *absvalues, size_t *bits); diff --git a/src/3rdparty/libjpeg/src/jsimd_none.c b/src/3rdparty/libjpeg/src/jsimd_none.c index 3cb6c80f8a..a25db73899 100644 --- a/src/3rdparty/libjpeg/src/jsimd_none.c +++ b/src/3rdparty/libjpeg/src/jsimd_none.c @@ -2,8 +2,9 @@ * jsimd_none.c * * Copyright 2009 Pierre Ossman <ossman@cendio.se> for Cendio AB - * Copyright (C) 2009-2011, 2014, D. R. Commander. - * Copyright (C) 2015-2016, 2018, Matthieu Darbois. + * Copyright (C) 2009-2011, 2014, 2022, D. R. Commander. + * Copyright (C) 2015-2016, 2018, 2022, Matthieu Darbois. + * Copyright (C) 2020, Arm Limited. * * Based on the x86 SIMD extension for IJG JPEG library, * Copyright (C) 1999-2006, MIYASAKA Masaru. @@ -169,6 +170,12 @@ jsimd_can_h2v1_fancy_upsample(void) return 0; } +GLOBAL(int) +jsimd_can_h1v2_fancy_upsample(void) +{ + return 0; +} + GLOBAL(void) jsimd_h2v2_fancy_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, JSAMPARRAY input_data, JSAMPARRAY *output_data_ptr) @@ -181,6 +188,12 @@ jsimd_h2v1_fancy_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, { } +GLOBAL(void) +jsimd_h1v2_fancy_upsample(j_decompress_ptr cinfo, jpeg_component_info *compptr, + JSAMPARRAY input_data, JSAMPARRAY *output_data_ptr) +{ +} + GLOBAL(int) jsimd_can_h2v2_merged_upsample(void) { @@ -399,7 +412,7 @@ jsimd_can_encode_mcu_AC_first_prepare(void) GLOBAL(void) jsimd_encode_mcu_AC_first_prepare(const JCOEF *block, const int *jpeg_natural_order_start, int Sl, - int Al, JCOEF *values, size_t *zerobits) + int Al, UJCOEF *values, size_t *zerobits) { } @@ -412,7 +425,7 @@ jsimd_can_encode_mcu_AC_refine_prepare(void) GLOBAL(int) jsimd_encode_mcu_AC_refine_prepare(const JCOEF *block, const int *jpeg_natural_order_start, int Sl, - int Al, JCOEF *absvalues, size_t *bits) + int Al, UJCOEF *absvalues, size_t *bits) { return 0; } diff --git a/src/3rdparty/libjpeg/src/jstdhuff.c b/src/3rdparty/libjpeg/src/jstdhuff.c index 036d6495a5..345b513d4d 100644 --- a/src/3rdparty/libjpeg/src/jstdhuff.c +++ b/src/3rdparty/libjpeg/src/jstdhuff.c @@ -4,7 +4,7 @@ * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1998, Thomas G. Lane. * libjpeg-turbo Modifications: - * Copyright (C) 2013, D. R. Commander. + * Copyright (C) 2013, 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -29,7 +29,7 @@ add_huff_table(j_common_ptr cinfo, JHUFF_TBL **htblptr, const UINT8 *bits, return; /* Copy the number-of-symbols-of-each-code-length counts */ - MEMCOPY((*htblptr)->bits, bits, sizeof((*htblptr)->bits)); + memcpy((*htblptr)->bits, bits, sizeof((*htblptr)->bits)); /* Validate the counts. We do this here mainly so we can copy the right * number of symbols from the val[] array, without risking marching off @@ -41,8 +41,9 @@ add_huff_table(j_common_ptr cinfo, JHUFF_TBL **htblptr, const UINT8 *bits, if (nsymbols < 1 || nsymbols > 256) ERREXIT(cinfo, JERR_BAD_HUFF_TABLE); - MEMCOPY((*htblptr)->huffval, val, nsymbols * sizeof(UINT8)); - MEMZERO(&((*htblptr)->huffval[nsymbols]), (256 - nsymbols) * sizeof(UINT8)); + memcpy((*htblptr)->huffval, val, nsymbols * sizeof(UINT8)); + memset(&((*htblptr)->huffval[nsymbols]), 0, + (256 - nsymbols) * sizeof(UINT8)); /* Initialize sent_table FALSE so table will be written to JPEG file. */ (*htblptr)->sent_table = FALSE; diff --git a/src/3rdparty/libjpeg/src/jutils.c b/src/3rdparty/libjpeg/src/jutils.c index 5c5bb17dc5..d86271624a 100644 --- a/src/3rdparty/libjpeg/src/jutils.c +++ b/src/3rdparty/libjpeg/src/jutils.c @@ -3,8 +3,8 @@ * * This file was part of the Independent JPEG Group's software: * Copyright (C) 1991-1996, Thomas G. Lane. - * It was modified by The libjpeg-turbo Project to include only code - * relevant to libjpeg-turbo. + * libjpeg-turbo Modifications: + * Copyright (C) 2022, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -110,7 +110,7 @@ jcopy_sample_rows(JSAMPARRAY input_array, int source_row, for (row = num_rows; row > 0; row--) { inptr = *input_array++; outptr = *output_array++; - MEMCOPY(outptr, inptr, count); + memcpy(outptr, inptr, count); } } @@ -120,7 +120,7 @@ jcopy_block_row(JBLOCKROW input_row, JBLOCKROW output_row, JDIMENSION num_blocks) /* Copy a row of coefficient blocks from one place to another. */ { - MEMCOPY(output_row, input_row, num_blocks * (DCTSIZE2 * sizeof(JCOEF))); + memcpy(output_row, input_row, num_blocks * (DCTSIZE2 * sizeof(JCOEF))); } @@ -129,5 +129,5 @@ jzero_far(void *target, size_t bytestozero) /* Zero out a chunk of memory. */ /* This might be sample-array data, block-array data, or alloc_large data. */ { - MEMZERO(target, bytestozero); + memset(target, 0, bytestozero); } diff --git a/src/3rdparty/libjpeg/src/jversion.h b/src/3rdparty/libjpeg/src/jversion.h index ab4a2c5703..ea6de648d9 100644 --- a/src/3rdparty/libjpeg/src/jversion.h +++ b/src/3rdparty/libjpeg/src/jversion.h @@ -2,9 +2,9 @@ * jversion.h * * This file was part of the Independent JPEG Group's software: - * Copyright (C) 1991-2012, Thomas G. Lane, Guido Vollbeding. + * Copyright (C) 1991-2020, Thomas G. Lane, Guido Vollbeding. * libjpeg-turbo Modifications: - * Copyright (C) 2010, 2012-2020, D. R. Commander. + * Copyright (C) 2010, 2012-2023, D. R. Commander. * For conditions of distribution and use, see the accompanying README.ijg * file. * @@ -30,23 +30,25 @@ * NOTE: It is our convention to place the authors in the following order: * - libjpeg-turbo authors (2009-) in descending order of the date of their * most recent contribution to the project, then in ascending order of the - * date of their first contribution to the project + * date of their first contribution to the project, then in alphabetical + * order * - Upstream authors in descending order of the date of the first inclusion of * their code */ #define JCOPYRIGHT \ - "Copyright (C) 2009-2020 D. R. Commander\n" \ - "Copyright (C) 2011-2016 Siarhei Siamashka\n" \ + "Copyright (C) 2009-2023 D. R. Commander\n" \ + "Copyright (C) 2015, 2020 Google, Inc.\n" \ + "Copyright (C) 2019-2020 Arm Limited\n" \ "Copyright (C) 2015-2016, 2018 Matthieu Darbois\n" \ + "Copyright (C) 2011-2016 Siarhei Siamashka\n" \ "Copyright (C) 2015 Intel Corporation\n" \ - "Copyright (C) 2015 Google, Inc.\n" \ + "Copyright (C) 2013-2014 Linaro Limited\n" \ "Copyright (C) 2013-2014 MIPS Technologies, Inc.\n" \ - "Copyright (C) 2013 Linaro Limited\n" \ + "Copyright (C) 2009, 2012 Pierre Ossman for Cendio AB\n" \ "Copyright (C) 2009-2011 Nokia Corporation and/or its subsidiary(-ies)\n" \ - "Copyright (C) 2009 Pierre Ossman for Cendio AB\n" \ "Copyright (C) 1999-2006 MIYASAKA Masaru\n" \ - "Copyright (C) 1991-2016 Thomas G. Lane, Guido Vollbeding" + "Copyright (C) 1991-2020 Thomas G. Lane, Guido Vollbeding" #define JCOPYRIGHT_SHORT \ - "Copyright (C) 1991-2020 The libjpeg-turbo Project and many others" + "Copyright (C) 1991-2023 The libjpeg-turbo Project and many others" |