simd/Makefile.am


Log

Author Commit Date CI Message
DRC 53c635b8 2016-02-08T14:03:13 Fix 'make dist'; Include LICENSE.md in packages
DRC f3a8684c 2016-01-07T00:19:43 SSE2 SIMD implementation of Huffman encoding Full-color compression speedups relative to libjpeg-turbo 1.4.2: 2.8 GHz Intel Xeon W3530, Linux, 64-bit: 2.2-18% (avg. 9.5%) 2.8 GHz Intel Xeon W3530, Linux, 32-bit: 10-25% (avg. 17%) 2.3 GHz AMD A10-4600M APU, Linux, 64-bit: 4.9-17% (avg. 11%) 2.3 GHz AMD A10-4600M APU, Linux, 32-bit: 8.8-19% (avg. 15%) 3.0 GHz Intel Core i7, OS X, 64-bit: 3.5-16% (avg. 10%) 3.0 GHz Intel Core i7, OS X, 32-bit: 4.8-14% (avg. 11%) 2.6 GHz AMD Athlon 64 X2 5050e: Performance-neutral (give or take a few percent) Full-color compression speedups relative to IPP: 2.8 GHz Intel Xeon W3530, Linux, 64-bit: 4.8-34% (avg. 19%) 2.8 GHz Intel Xeon W3530, Linux, 32-bit: -19%-7.0% (avg. -7.0%) Refer to #42 for discussion. Numerous other approaches were attempted, but this one proved to be the most performant across all platforms. This commit also fixes #3 (works around, really-- the clang-compiled version of jchuff.c still performs 20% worse than its GCC-compiled counterpart, but that code is now bypassed by the new SSE2 Huffman algorithm.) Based on: https://github.com/mayeut/libjpeg-turbo/commit/2cb4d41330e1edc4469f6b97ba73b73abfbeb02f https://github.com/mayeut/libjpeg-turbo/commit/36c94e050d117912adbff9fbcc6fe307df240168
DRC 691cd933 2015-06-20T16:36:32 Fix 'make dist' git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1574 632fc199-4ca6-4c93-a231-07263d6284db
DRC 89b5e06d 2015-06-20T16:20:53 Studies show that GCC v5.1.0 performs as well as or better than v4.2, but v4.7.x-v4.9.x do not perform as well as v4.2. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1571 632fc199-4ca6-4c93-a231-07263d6284db
DRC 3b7015d5 2015-02-23T19:03:29 Enable silent build rules for the NASM objects, if the source is configured with automake 1.11 or later. NOTE: the build still spits out "error: ignoring unknown tag NASM" for each object, but unfortunately, if we remove "--tag NASM" from the command line, the build breaks under older versions of automake (it aborts with "unable to infer tagged configuration.") git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/branches/1.4.x@1534 632fc199-4ca6-4c93-a231-07263d6284db
DRC 86af36ae 2015-01-14T13:27:32 AltiVec SIMD implementation of H2V1 and H2V2 merged upsampling git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1503 632fc199-4ca6-4c93-a231-07263d6284db
DRC 52a4ec6c 2015-01-13T09:02:29 AltiVec SIMD implementation of H2V1 and H2V2 fancy upsampling git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1495 632fc199-4ca6-4c93-a231-07263d6284db
DRC ac4daa77 2015-01-10T22:56:26 AltiVec SIMD implementation of YCC-to-RGB color conversion git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1489 632fc199-4ca6-4c93-a231-07263d6284db
DRC 22048207 2015-01-08T06:18:33 AltiVec SIMD implementation of 2x1 and 2x2 downsampling git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1483 632fc199-4ca6-4c93-a231-07263d6284db
DRC 577ecd93 2014-12-23T04:14:54 AltiVec SIMD implementation of sample conversion and integer quantization git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1474 632fc199-4ca6-4c93-a231-07263d6284db
DRC b1fec4ff 2014-12-22T14:10:33 AltiVec SIMD implementation of RGB-to-Grayscale color conversion git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1471 632fc199-4ca6-4c93-a231-07263d6284db
DRC 5976e425 2014-12-22T13:57:30 Remove unneeded code; Make sure jccolor-altivec.o will be rebuilt if jccolext-altivec.c changes. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1470 632fc199-4ca6-4c93-a231-07263d6284db
DRC 62bae204 2014-12-22T13:42:26 AltiVec SIMD implementation of RGB-to-YCC color conversion git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1469 632fc199-4ca6-4c93-a231-07263d6284db
DRC 535674b1 2014-12-22T01:00:42 Split AltiVec algorithms into separate files for ease of maintenance; Rename constants using lowercase so they are not confused with macros git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1463 632fc199-4ca6-4c93-a231-07263d6284db
DRC 7affbfc2 2014-09-05T07:23:12 The AltiVec code actually works on 32-bit PowerPC platforms as well, so change the "powerpc64" token to "powerpc". Also clean up the shift code, which wasn't building properly on OS X. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1406 632fc199-4ca6-4c93-a231-07263d6284db
DRC cd2d8e1c 2014-09-05T06:33:42 AltiVec SIMD implementation of fast forward DCT git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1405 632fc199-4ca6-4c93-a231-07263d6284db
DRC 0a9a2526 2014-08-29T01:53:17 Rename the ARM64 assembly file to match the C file git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1390 632fc199-4ca6-4c93-a231-07263d6284db
DRC 495e4342 2014-05-19T19:13:22 Allow for building the MIPS DSPr2 extensions if the host is mips-* as well as mipsel-*. The DSPr2 extensions are little endian, so we still have to check that the compiler defines __MIPSEL__ before enabling them. This paves the way for supporting big-endian MIPS, and in the near term, it allows the SIMD extensions to be built with Sourcery CodeBench. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1316 632fc199-4ca6-4c93-a231-07263d6284db
DRC b7753510 2014-05-11T09:36:25 Convert tabs to spaces in the libjpeg code and the SIMD code (TurboJPEG retains the use of tabs for historical reasons. They were annoying in the libjpeg code primarily because they were not consistently used and because they were used to format as well as indent the code. In the case of TurboJPEG, tabs are used just to indent the code, so even if the editor assumes a different tab width, the code will still be readable.) git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/branches/1.3.x@1285 632fc199-4ca6-4c93-a231-07263d6284db
DRC 24e92e9f 2014-05-10T09:53:34 Using subdirectories unfortunately opened up a can of worms. In order to prevent object name conflicts, it is necessary to use the subdir-objects automake directive, but it simply doesn't work right on some of the versions of automake we still have to support. Another option would be to add a separate Makefile.am file to each subdirectory, but that requires maintaining a completely different set of build rules for each one. Fortunately, however, we're in the 21st century now, so we can use filenames longer than 8.3. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1282 632fc199-4ca6-4c93-a231-07263d6284db
DRC 72130be9 2014-05-09T20:14:26 Re-organize the x86/x86-64 SIMD routines into separate folders by instruction set so we can name each routine similarly to its corresponding C file. This also makes it easier to add support for new instruction sets. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1280 632fc199-4ca6-4c93-a231-07263d6284db
DRC 1a45b81f 2014-05-09T18:06:58 Remove trailing spaces (+ one additional tab in TJUnitTest.java that was missed in the previous commit) git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1279 632fc199-4ca6-4c93-a231-07263d6284db
DRC 2d07ee51 2014-02-05T19:03:41 Create a separate stub file for 64-bit ARM, since it currently implements only the decompression-related functions. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1109 632fc199-4ca6-4c93-a231-07263d6284db
DRC ba55b2cd 2014-02-05T08:15:44 First pass at ARMv8 64-bit NEON SIMD support git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1108 632fc199-4ca6-4c93-a231-07263d6284db
DRC 371b420e 2013-08-23T07:57:21 Fix 'make dist' git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1024 632fc199-4ca6-4c93-a231-07263d6284db
DRC 0be9fa57 2013-07-24T21:50:20 SIMD support for performing color conversion using MIPS DSPr2 instructions git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@993 632fc199-4ca6-4c93-a231-07263d6284db
DRC 1ca924a5 2011-11-29T08:58:27 NASM automatically adds the current directory to the include path, but YASM doesn't, so we need to explicitly add it. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@724 632fc199-4ca6-4c93-a231-07263d6284db
DRC 321e0686 2011-05-03T08:47:43 ARM NEON support git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@607 632fc199-4ca6-4c93-a231-07263d6284db
DRC e856eb0f 2011-03-04T14:05:23 Include new grayscale SIMD routines in dist git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@508 632fc199-4ca6-4c93-a231-07263d6284db
DRC c8666333 2011-02-18T11:23:45 SIMD-accelerated RGB-to-Grayscale color conversion git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@393 632fc199-4ca6-4c93-a231-07263d6284db
DRC 1f80a10b 2010-10-18T00:15:31 Allow RPMs and SRPMs to be generated with v7/v8b emulation; Add CMake scripts and other missing files to source distribution git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@276 632fc199-4ca6-4c93-a231-07263d6284db
DRC 8a26fdea 2010-02-16T22:34:12 Fix build with Sun Studio git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@113 632fc199-4ca6-4c93-a231-07263d6284db
DRC 327184fb 2010-02-14T02:15:52 builddir isn't always defined git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@89 632fc199-4ca6-4c93-a231-07263d6284db
DRC 971d8518 2010-02-13T22:55:05 Enable (i.e. fix) VPATH and out-of-tree builds git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@87 632fc199-4ca6-4c93-a231-07263d6284db
DRC 4f442818 2010-02-05T09:19:13 Oops. Use EGREP instead of GREP, because GREP is not available in earlier versions of autotools git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@75 632fc199-4ca6-4c93-a231-07263d6284db
DRC 2a46aeec 2010-01-29T07:14:57 Fix build problems on Solaris 10 git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@74 632fc199-4ca6-4c93-a231-07263d6284db
Adam Tkac 0ba6826d 2009-09-14T13:07:14 Merge unified_buildsys branch to trunk (r3892 - r3898) git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@62 632fc199-4ca6-4c93-a231-07263d6284db
Pierre Ossman f1418c76 2009-07-10T12:47:13 Make sure the indirectly used assembly files get packaged. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@52 632fc199-4ca6-4c93-a231-07263d6284db
Pierre Ossman ba82ddf6 2009-06-29T11:20:42 Clean up SIMD glue code The SIMD glue code has gotten a bit #ifdef heavy so clean it up by having one file for each possible SIMD arch. This also allows a simplification of the x86_64 code as SSE/SSE2 is always known to exist on that arch. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@49 632fc199-4ca6-4c93-a231-07263d6284db
DRC cdc8ac3e 2009-06-25T20:38:31 64-bit SIMD acceleration git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@45 632fc199-4ca6-4c93-a231-07263d6284db
DRC 3a631844 2009-06-25T20:33:51 Fix dependencies for merged upsampling code git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@44 632fc199-4ca6-4c93-a231-07263d6284db
DRC f25c071e 2009-04-03T12:00:51 Implement new colorspaces to allow directly compressing from/decompressing to RGB/RGBX/BGR/BGRX/XBGR/XRGB without conversion git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@35 632fc199-4ca6-4c93-a231-07263d6284db
Pierre Ossman 3a65ef47 2009-03-16T13:34:18 Work around subdirectory bug in older automake. Older versions of automake doesn't properly support no-recursive make. Reimplement the build system by having a local Makefile.am in the simd/ directory. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@31 632fc199-4ca6-4c93-a231-07263d6284db