6eb7d3798b5a79347c62825fc4c16f7ce673bdd0 |
|
16-Oct-2016 |
Alex Naidis <alex.naidis@linux.com> |
libjpeg-turbo: Upgrade to 1.5.1 The changes from 1.4.2 to 1.5.1 include a big amount of fixes and huge performance improvements. As highlights there is a full ARM 64-bit (ARMv8) NEON SIMD implementation which improves compression of full-color JPEGs by about 2-2.5x on average on Cortex-A53 and Cortex-A57 cores. Also SIMD acceleration for Huffman encoding on NEON-capable ARM 32-bit and 64-bit platforms was added. Performance on x86/x86_64 was also improved by adding better optimized SSE2 routines. For the full changelog, please see the ChangeLog.md file. Partial decoding optimizations, the security fix to adress b/27494207 and the fix for the AARCH64 conformance issueare present in the release. The README.android file was edited to reflect this. The configuration files were regenerated and all Android specific changes were applied. BUG:28268702 Change-Id: I538291d894df1da01d3f733771647df1fb61ec42 Signed-off-by: Alex Naidis <alex.naidis@linux.com>
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
d729f4da9c86b7212912a7d59e49d061d0e61d5f |
|
23-Aug-2014 |
DRC <dcommander@users.sourceforge.net> |
ARM NEON SIMD support for YCC-to-RGB565 conversion, and optimizations to the existing YCC-to-RGB conversion code: ----- https://github.com/ssvb/libjpeg-turbo/commit/aee36252be20054afce371a92406fc66ba6627b5.patch From aee36252be20054afce371a92406fc66ba6627b5 Mon Sep 17 00:00:00 2001 From: Siarhei Siamashka <siarhei.siamashka@gmail.com> Date: Wed, 13 Aug 2014 03:50:22 +0300 Subject: [PATCH] ARM: Faster NEON yuv->rgb conversion for Krait and Cortex-A15 The older code was developed and tested only on ARM Cortex-A8 and ARM Cortex-A9. Tuning it for newer ARM processors can introduce some speed-up (up to 20%). The performance of the inner loop (conversion of 8 pixels) improves from ~27 cycles down to ~22 cycles on Qualcomm Krait 300, and from ~20 cycles down to ~18 cycles on ARM Cortex-A15. The performance remains exactly the same on ARM Cortex-A7 (~58 cycles), ARM Cortex-A8 (~25 cycles) and ARM Cortex-A9 (~30 cycles) processors. Also use larger indentation in the source code for separating two independent instruction streams. ----- https://github.com/ssvb/libjpeg-turbo/commit/a5efdbf22ce9c1acd4b14a353cec863c2c57557e.patch From a5efdbf22ce9c1acd4b14a353cec863c2c57557e Mon Sep 17 00:00:00 2001 From: Siarhei Siamashka <siarhei.siamashka@gmail.com> Date: Wed, 13 Aug 2014 07:23:09 +0300 Subject: [PATCH] ARM: NEON optimized yuv->rgb565 conversion The performance of the inner loop (conversion of 8 pixels): * ARM Cortex-A7: ~55 cycles * ARM Cortex-A8: ~28 cycles * ARM Cortex-A9: ~32 cycles * ARM Cortex-A15: ~20 cycles * Qualcomm Krait: ~24 cycles Based on the Linaro rgb565 patch from https://sourceforge.net/p/libjpeg-turbo/patches/24/ but implements better instructions scheduling. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1385 632fc199-4ca6-4c93-a231-07263d6284db
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
1419852c42a0b8d8779fa89419cf2824b9eb5866 |
|
15-May-2014 |
DRC <dcommander@users.sourceforge.net> |
Clean up code formatting in the SIMD interface functions git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@1305 632fc199-4ca6-4c93-a231-07263d6284db
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
1a45b81fa2edf682ae72251a18fd84eacef9dbbe |
|
09-May-2014 |
DRC <dcommander@users.sourceforge.net> |
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
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
67ce3b2352fe1f7511edbfed74ec6960e41e97dc |
|
19-Dec-2011 |
DRC <dcommander@users.sourceforge.net> |
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. git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@732 632fc199-4ca6-4c93-a231-07263d6284db
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
392e0483a33f04efd0a68512ab6767568bccb5c4 |
|
18-Feb-2011 |
DRC <dcommander@users.sourceforge.net> |
Updated (C) git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@394 632fc199-4ca6-4c93-a231-07263d6284db
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
c86663335ba74e0c732f1015f03181b327646e8d |
|
18-Feb-2011 |
DRC <dcommander@users.sourceforge.net> |
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
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
af1ca9bcc1674e0c6d07127f7dc04b138f8b1c98 |
|
02-Feb-2011 |
DRC <dcommander@users.sourceforge.net> |
Clarify that the C wrappers and headers fall under the same license as the rest of the SIMD code git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/branches/1.0.x@335 632fc199-4ca6-4c93-a231-07263d6284db
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
c06073a94954788212fc23f699a96132f0d3d201 |
|
06-Sep-2010 |
DRC <dcommander@users.sourceforge.net> |
Remove simd/ prefix from #include (not necessary and was causing problems with Visual Studio project) git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@228 632fc199-4ca6-4c93-a231-07263d6284db
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
30959719ad8f932a1719f5c09af6a84ce1f519a3 |
|
07-Aug-2010 |
DRC <dcommander@users.sourceforge.net> |
Fix typo in SIMD dispatch routines which was causing 4:2:0 upsampling to be used instead of 4:2:2 when decompressing JPEG images using SSE2 code git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@223 632fc199-4ca6-4c93-a231-07263d6284db
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
048990944fccc2602c892dabace774f40271a61e |
|
27-Feb-2010 |
DRC <dcommander@users.sourceforge.net> |
Bleepin' Windows uses LLP64, not LP64 git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@158 632fc199-4ca6-4c93-a231-07263d6284db
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
321ad5131492107e0e501f13a37fb3ac7f36f06d |
|
08-Oct-2009 |
DRC <dcommander@users.sourceforge.net> |
Enable 64-bit build on Snow Leopard git-svn-id: svn+ssh://svn.code.sf.net/p/libjpeg-turbo/code/trunk@68 632fc199-4ca6-4c93-a231-07263d6284db
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|
ba82ddf6ca0270d65e3b8fb56b2d3116cf423709 |
|
29-Jun-2009 |
Pierre Ossman <ossman@cendio.se> |
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
/external/libjpeg-turbo/simd/jsimd_x86_64.c
|