- b87a0b4 Fix the x86 build with NASM 0.98. Since NASM 0.98 is the default version on OS X, we want to at least allow people to build 32-bit code with it, even though it can't properly build 64-bit code. by DRC · 12 years ago
- 29d8f25 Fix build issues that occurred whenever the source directory contained the letters "col", "mer", or "gra". by DRC · 12 years ago
- 112a0bb More recent versions of autoconf add -traditional-cpp to the CPP flags, which causes jsimdcfg.inc.h to not preprocess correctly unless we expand all of the instances of the #definev macro. by DRC · 12 years ago
- 8126d0c Fixed regression caused by a bug in the 32-bit strict memory access code in jdmrgss2.asm (contributed by Chromium to stop valgrind from whining whenever the output buffer size was not evenly divisible by 16 bytes.) On Linux/x86, this regression generated incorrect pixels on the right-hand side of images whose rows were not 16-byte aligned, whenever fancy upsampling was used. This patch also enables the strict memory access code on all platforms, not just Linux (it does no harm on other platforms) and removes a couple of pcmpeqb instructions that were rendered unnecessary by r835. by DRC · 12 years ago
- 316617f Accelerated 4:2:2 upsampling routine for ARM (improves performance ~20-30% when decompressing 4:2:2 JPEGs using fancy upsampling) by DRC · 12 years ago
- 6979927 Eliminate the use of the MASKMOVDQU instruction, to speed up decompression performance by 10x on AMD Bobcat embedded processors (and ~5% on AMD desktop processors.) by DRC · 12 years ago
- 4f24016 Preserve all 128 bits of xmm6 and xmm7 by DRC · 12 years ago
- 8015a30 Visual Studio 2010 doesn't like the wildcard at compile time, so let CMake expand it instead. by DRC · 12 years ago
- d65d99a Compiler warnings by DRC · 13 years ago
- 67ce3b2 Added new alpha channel colorspace constants/pixel formats, so applications can specify that they need the unused byte in a 4-component RGB output buffer set to 0xFF when decompressing. by DRC · 13 years ago
- ebfe9e4 Fixed non-fatal out-of-bounds read in SSE2 SIMD code reported by valgrind when decompressing a JPEG image to a bitmap buffer whose size was not a multiple of 16 bytes. by DRC · 13 years ago
- c412184 Fixed non-fatal out-of-bounds read in SSE2 SIMD code reported by valgrind when decompressing a JPEG image to a bitmap buffer whose size was not a multiple of 16 bytes. by DRC · 13 years ago
- 795e6ad Fixed non-fatal out-of-bounds read in SSE2 SIMD code reported by valgrind when decompressing a JPEG image to a bitmap buffer whose size was not a multiple of 16 bytes. by DRC · 13 years ago
- 105f9a9 Expose NASM variable in ccmake by DRC · 13 years ago
- 0f905d3 Expose NASM variable in ccmake by DRC · 13 years ago
- 1ca924a NASM automatically adds the current directory to the include path, but YASM doesn't, so we need to explicitly add it. by DRC · 13 years ago
- c08e8c1 When decompressing to a 4-byte RGB buffer, set the unused byte to 0xFF so it can be interpreted as an opaque alpha channel. by DRC · 13 years ago
- b4570bb Improve performance of non-SIMD color conversion routines and use global constants to define colorspace extension parameters by DRC · 13 years ago
- b071f01 Update Nokia contact info by DRC · 13 years ago
- ad6955d Improve performance of IFAST iDCT by changing the order of transpose and descale steps by DRC · 13 years ago
- 5129e39 Make ARM ISLOW iDCT faster on typical cases, and eliminate the possibility of 16-bit overflows when handling arbitrary coefficients. by DRC · 13 years ago
- 98a44fe Improve the performance of YCbCr to RGB conversion on ARM by DRC · 13 years ago
- ce4e3e8 NEON-accelerated slow integer inverse DCT by DRC · 13 years ago
- 82bd521 NEON-accelerated quantization by DRC · 13 years ago
- 4b024a6 Improve performance of ARM NEON IFAST iDCT by DRC · 13 years ago
- 7a9376c ARM NEON-accelerated RGB-to-YCbCr conversion by DRC · 13 years ago
- b740054 Support for accelerated forward DCT using ARM NEON instructions by DRC · 13 years ago
- 8c60d22 NEON-optimized 2x2 and 4x4 scaled iDCTs by DRC · 13 years ago
- 4346f91 iOS ARM support by DRC · 13 years ago
- e19f15e Not necessary to save r10 and r11, since these are scratch registers by DRC · 13 years ago
- 44a97e8 Not necessary to save r10 and r11, since these are scratch registers by DRC · 13 years ago
- 321e068 ARM NEON support by DRC · 13 years ago
- 56fb237 YASM support by DRC · 13 years ago
- dc6f6a9 Don't need MSVC definition in assembler code anymore by DRC · 13 years ago
- 577ae78 Don't need MSVC definition in assembler code anymore by DRC · 13 years ago
- 926e01f Attempt to make build work with Visual Studio 2010 (still doesn't work if SIMD or Java are enabled-- see bug tracker) by DRC · 13 years ago
- b05ebcb Attempt to make build work with Visual Studio 2010 (still doesn't work if SIMD or Java are enabled-- see bug tracker) by DRC · 13 years ago
- 58b6ec4 Ensure that *gra*.asm is rebuilt whenever *gry*.asm changes. by DRC · 13 years ago
- 1da7858 MinGW64 no longer prefixes symbols with underscores by default by DRC · 13 years ago
- cd2e694 MinGW64 no longer prefixes symbols with underscores by default by DRC · 13 years ago
- 47aaf6f MinGW64 no longer prefixes symbols with underscores by default by DRC · 13 years ago
- e856eb0 Include new grayscale SIMD routines in dist by DRC · 13 years ago
- da5220a Fix Windows build; Add grayscale tests to Windows build; Add proper implementation of snprintf on Windows and use snprintf() in TurboJPEG library and tests instead of sprintf() by DRC · 14 years ago
- ddb158c Add short names for RGB->grayscale MMX functions by DRC · 14 years ago
- 07122dd Change filename in header by DRC · 14 years ago
- 392e048 Updated (C) by DRC · 14 years ago
- c866633 SIMD-accelerated RGB-to-Grayscale color conversion by DRC · 14 years ago
- af1ca9b Clarify that the C wrappers and headers fall under the same license as the rest of the SIMD code by DRC · 14 years ago
- daae6d9 Clarify that the C wrappers and headers fall under the same license as the rest of the SIMD code by DRC · 14 years ago
- 1f80a10 Allow RPMs and SRPMs to be generated with v7/v8b emulation; Add CMake scripts and other missing files to source distribution by DRC · 14 years ago
- d830d36 Fix dependencies for the color conversion routines, which build other .asm files indirectly by DRC · 14 years ago
- c869c2c Unix LF by DRC · 14 years ago
- cd5f2ee Allow nasm path to be overridden by DRC · 14 years ago
- 8469703 First pass at a CMake build system by DRC · 14 years ago
- c06073a Remove simd/ prefix from #include (not necessary and was causing problems with Visual Studio project) by DRC · 14 years ago
- 3095971 Fix typo in SIMD dispatch routines which was causing 4:2:0 upsampling to by DRC · 14 years ago
- 231a43f Fix build with YASM by DRC · 14 years ago
- b6f097f xmm6-xmm15 must also be preserved by the callee; Win64 should fully work now by DRC · 14 years ago
- d3e320f Win64 should fully work now by DRC · 14 years ago
- 4132b1d Updated (C) by DRC · 14 years ago
- e728ed7 Win64 (mostly) works now by DRC · 14 years ago
- a644fb0 Use correct filenames in headers by DRC · 14 years ago
- fb096ff Only have to use 32-bit mov for r14, since it is populated from the stack by DRC · 14 years ago
- 14ecb53 Attempt to fix Win64 issues by DRC · 14 years ago
- f8b77c4 Don't allow code path to be forced on a system that doesn't support it by DRC · 14 years ago
- 0489909 Bleepin' Windows uses LLP64, not LP64 by DRC · 15 years ago
- 0c6a271 Make TJ_FORCE* options work correctly by DRC · 15 years ago
- 26b208d MinGW64 requires that the functions be prefixed with an underscore. Visual C++ apparently doesn't by DRC · 15 years ago
- 8b014d7 First attempt at Win64 support by DRC · 15 years ago
- c932768 Oops. Leftover bugs from the 64-bit port (cause seg fault when using JDCT_FLOAT or JDCT_IFAST when decompressing) by DRC · 15 years ago
- 8a26fde Fix build with Sun Studio by DRC · 15 years ago
- 327184f builddir isn't always defined by DRC · 15 years ago
- 971d851 Enable (i.e. fix) VPATH and out-of-tree builds by DRC · 15 years ago
- 4f44281 Oops. Use EGREP instead of GREP, because GREP is not available in earlier versions of autotools by DRC · 15 years ago
- 2a46aee Fix build problems on Solaris 10 by DRC · 15 years ago
- e54fb0b [Bugfix] Mark stack as non-executable in NASM code. by Adam Tkac · 15 years ago
- 321ad51 Enable 64-bit build on Snow Leopard by DRC · 15 years ago
- 132b5fd Work around alignment issue with OS X linker which was causing problems on Leopard and Snow Leopard by DRC · 15 years ago
- 0ba6826 Merge unified_buildsys branch to trunk (r3892 - r3898) by Adam Tkac · 15 years ago
- adfd233 Windows build fixes: by Peter Åstrand · 15 years ago
- 62aaa7e Oops. Clean up a couple of remaining 32-bit memory accesses by DRC · 15 years ago
- f1418c7 Make sure the indirectly used assembly files get packaged. by Pierre Ossman · 15 years ago
- 9a6b8dc Make x86_64 SIMD code PIC friendly by Pierre Ossman · 15 years ago
- ba82ddf Clean up SIMD glue code by Pierre Ossman · 15 years ago
- cdc8ac3 64-bit SIMD acceleration by DRC · 15 years ago
- 3a63184 Fix dependencies for merged upsampling code by DRC · 15 years ago
- 720e161 Add colorspace extensions to merged upsampling routines by DRC · 15 years ago
- 16ed1f5 Oops. jcolsamp.inc should be included from j[c|d]clr[mmx|ss2].asm so it received the effect of the redefined RGB_* macros by DRC · 15 years ago
- a2d19e1 Oops. jcolsamp.inc needs to be included from j[c|d]clr[mmx|ss2].asm so that it receives the effect of the redefined RGB_* macros by DRC · 15 years ago
- f25c071 Implement new colorspaces to allow directly compressing from/decompressing to RGB/RGBX/BGR/BGRX/XBGR/XRGB without conversion by DRC · 15 years ago
- c7fbf29 Fix build for RGB_PIXELSIZE==4 by DRC · 15 years ago
- 3a65ef4 Work around subdirectory bug in older automake. by Pierre Ossman · 15 years ago
- eea7215 Add SSE2 SIMD implementation of computationally intensive routines. by Pierre Ossman · 15 years ago
- 018fc42 Add SSE SIMD implementation of computationally intensive routines. by Pierre Ossman · 15 years ago
- 65d0317 Add 3DNow SIMD implementation of computationally intensive routines. by Pierre Ossman · 15 years ago
- 5eb84ff Add MMX SIMD implementation of computationally intensive routines. by Pierre Ossman · 15 years ago
- 2ae181c Implement x86 SIMD framework by Pierre Ossman · 15 years ago