Commit Graph

82 Commits

Author SHA1 Message Date
skal
97e5fac389 add some colorspace conversion functions in NEON
new file: lossless_neon.c
speedup is ~5%

gcc 4.6.3 seems to be doing some sub-optimal things here,
storing register on stack using 'vstmia' and such.
Looks similar to gcc.gnu.org/bugzilla/show_bug.cgi?id=51509

I've tried adding  -fno-split-wide-types and it does help
the generated assembly. But the overall speed gets worse with
this flag. We should only compile lossless_neon.c with it -> urk.

Change-Id: I2ccc0929f5ef9dfb0105960e65c0b79b5f18d3b0
2014-03-31 17:47:46 +02:00
skal
0f4f721b12 separate SSE2 lossless functions into its own file
expose the predictor array as function pointers instead
of each individual sub-function

+ merged Average2() into ClampedAddSubtractHalf directly
+ unified the signature as "VP8LProcessBlueAndRedFunc"

no speed diff observed

Change-Id: Ic3c45dff11884a8330a9ad38c2c8e82491c6e044
2014-03-27 21:43:55 +01:00
skal
8992ddb756 use static clipping tables
(shared with mips32)
removed abs1[] table along the way

sub-1% speed-up, but still...

Change-Id: I8c29a8a0285076cb3423b01ffae9fcc465da6a81
2014-02-13 19:32:59 -08:00
Djordje Pesut
dd438c9a7d MIPS: MIPS32r1: Optimization of some simple point-sampling functions. PATCH [6/6]
Change-Id: I2020e71e9be5d17d4bf67cabf6c470ca43d5d838
2014-01-29 15:37:31 +01:00
Djordje Pesut
04336fc7f8 MIPS: MIPS32r1: Optimization of function TransformOne. PATCH [4/6]
Change-Id: I5b98e2de940977538cf91bfa2128f4d1daa5c170
2014-01-28 20:10:43 -08:00
James Zern
ad26df1a05 makefile.unix: clean up libgif2webp_util.a
Change-Id: Iee8650ce91f65f2db568926773d44441633459f3
2013-12-18 23:26:18 -08:00
James Zern
5009b22746 makefile.unix: install binaries in $(DESTDIR)/bin/
Change-Id: I6c8f87aaa185fc043b80800430d99deaa1e5966b
2013-12-06 20:40:12 -08:00
Pascal Massimino
98aa33cf1e extract random utils to their own file util/random.[ch]
they'll be used for decoding too, probably.

Change-Id: Id9cbb250c74fc0e876d4ea46b1b3dbf8356d6725
2013-10-30 02:00:33 -07:00
Pascal Massimino
79ff03460a add proper WEBP_HAVE_GIF and WEBP_HAVE_GL flags
...and make gif2webp and vwebp compile without them.

Makefile.vc still to be updated...
Meanwhile the CL environment variable can be supplemented with

set CL=/DWEBP_HAVE_GL /IC:\opt\freeglut\include

Change-Id: I37a60b8c32aafd125bffa98b6cc9f57c022ebbd0
2013-09-13 12:46:41 -07:00
Urvang Joshi
d538cea8c2 gif2webp: Support a 'min' and 'max' key frame interval
Change-Id: If910c8849d7210e7227db0f45affa26fce7da102
2013-09-12 13:41:09 -07:00
Pascal Massimino
8dcae8b3cf fix rescaling-with-alpha inaccuracy
(still missing YUVA decoding case for now)

https://code.google.com/p/webp/issues/detail?id=160

Change-Id: If723b4a5c0a303d0853ec9d839f995adce056095
2013-07-26 12:10:26 -07:00
Urvang Joshi
068db59e26 Intertwined decoding of alpha and RGB
This will reduce the time to first decoded pixel.

Change-Id: I07b900c0ed4af3aac806b2731e11cd18ec16d016
2013-06-12 14:10:59 -07:00
James Zern
0e513f7ae3 configure: add GLUT detection; build vwebp
Change-Id: I7f0964db2d04c22ff9ec274e8cd1cbed7379a165
2013-05-07 11:42:15 -07:00
James Zern
289bc47b65 libwebp{demux,mux}: install mux_types.h
required by demux.h and mux.h

Change-Id: Ib33cee1d95e575233204e70b62caf7a6366772c7
2013-03-19 14:54:04 -07:00
James Zern
701b9e2af4 makefile.unix: dist related changes
* make the 'all' target really build everything (default is still the
    core examples).
  * add demux/mux.h to HDRS_INSTALLED, install the corresponding libs
    too
  * install vwebp, webpmux, gif2webp and related manpages

Change-Id: Ib6036f2a1a05e40f106914c4bdbe9e3ad7336464
2013-03-18 11:10:02 -07:00
skal
5cf7792e40 split quant_levels.c into decoder and encoder version
-> split libraries further into decoder / encoder
-> add libwebpdecoder.a in Makefile.unix
-> make dwebp link against libwebpdecoder.a in Makefile.unix

also: in makefile.unix, pass EXTRA_FLAGS to LDFLAGS too
(otherwise, -m32 wouldn't work, e.g.)

Change-Id: Ief3da02a729dd86bbaf949ed048836716941657f
2013-02-24 21:40:39 +01:00
Mans Rullgard
090b708a00 NEON optimised yuv to rgb conversion
- along the lines of the SSE chroma upsampling.
Total speedup is ~30%.

4% speed loss on YuvToRgbXX conversion using tables instead
of 14-bit fixed precision. TODO(later): investigate, and compare
to x86.

see http://code.google.com/p/webp/issues/detail?id=134

Change-Id: Idc2261037cd13b4553ca20ecc4c4007099c37009
2013-01-25 15:46:40 -08:00
James Zern
2b252a53a8 Merge "Provide option to swap bytes for 16 bit colormodes" 2013-01-22 15:00:39 -08:00
Vikas Arora
94a48b4bc3 Provide option to swap bytes for 16 bit colormodes
Color modes: RGB_565 & RGBA_4444
Change-Id: I571b6832b9848e5c4109272978f68623ca373383
2013-01-22 14:51:20 -08:00
skal
0d19fbff51 remove some -Wshadow warnings
these are quite noisy, but it's not a big deal to remove
them.

Change-Id: I5deb08f10263feb77e2cc8a70be44ad4f725febd
2013-01-22 23:06:28 +01:00
James Zern
6bf208748c add examples/metadata.c
relocates the static functions from metadata.h and adds MetadataCopy()
to the interface

Change-Id: I28bfa9233d3dd70dddf6b561fe0bf4be378db1ec
2012-12-11 18:10:09 -08:00
James Zern
1c1c5646a5 cwebp: extract tiff decoding to its own module
Change-Id: If98d0d37de34b63ac10d826150237b5f99446532
2012-12-10 22:37:04 -08:00
James Zern
6a871d66a4 cwebp: extract jpeg decoding to its own module
Change-Id: I45e1f0fa7b34286dd98926e0485e5a8ab1964570
2012-12-07 20:12:46 -08:00
James Zern
2ee228f9a5 cwebp: extract png decoding to its own module
Change-Id: I80dd09fe1f1c6317e9d0cdd422438bda11b43ed5
2012-12-07 20:02:41 -08:00
skal
657f5c91b1 move token buffer to its own file (token.c)
Change-Id: Ib9791c52f48d98fad5ed3830f36894ef5ac362fa
2012-12-03 13:50:14 +01:00
Urvang Joshi
2e7f6e8ef2 makefile.unix: Dependency on libraries
The binaries should have dependency on related libraries only.
For example, 'vwebp' sould not depend on libpng, libjpeg etc.

Change-Id: I517b7ae6a9092d5de891b46d0e75d07d91548e93
2012-11-21 12:12:29 -08:00
Urvang Joshi
23782f95b4 Separate out mux and demux code and libraries:
- Separate out mux.h and demux.h
- muxtypes.h: new header for data types common to mux/demux
- Move some misc read/write utilities to utils/utils.h
- Remove some duplicate methods.
- Separate out mux/demux libraries

Change-Id: If9b9569b10d55d922ad9317ef51710544315d6de
2012-11-19 11:40:18 -08:00
Urvang Joshi
11b2721258 Write a GIF to WebP converter based on libgif.
Change-Id: Ic64bf5307996a788ec61caf59a99ab37b4b1ae96
2012-11-08 11:21:25 -08:00
Pascal Massimino
75e5f17e3b ARM/NEON: 30% encoding speed-up
(implements the backward and forward transforms in the encoder)

original patch by Wayne Chen (datoudatou at gmail dot com)

Change-Id: Ic00f3bffcdf7a924f043006728735c810ee47a57
2012-10-31 14:00:20 -07:00
Pascal Massimino
80cc7303ab WebPCheckMalloc() and WebPCheckCalloc():
safe size-checking versions of malloc() and calloc()

Change-Id: Iffa3138c48b9b254b3d7eaad913e1f852d9dafba
2012-07-31 16:56:39 -07:00
James Zern
7415ae1386 makefile.unix: provide examples/webpmux target
this target is experimental so must be explicitly set

missing since:
e41a759 build: remove libwebpmux from default targets/config

Change-Id: I128d9e6a06ba3e7102ae3175dc3db52c6f4b1439
2012-07-19 14:39:26 -07:00
James Zern
d5e5ad6356 move decode_vp8.h from webp/ to dec/
the functions contained in it are now private

Change-Id: Ief6c81b32ae3f6d97052edac625716e5b909e66e
2012-07-16 22:12:59 -07:00
James Zern
8c1cc6b5dc makefile.unix dist: explicitly name installed includes
avoids mux.h, format_constants.h

Change-Id: I6c8458dd1ce7faa23fbf1e3df84a8dfc4a884a3e
2012-07-16 16:32:15 -07:00
James Zern
e41a759615 build: remove libwebpmux from default targets/config
the extended file format is still under development and related
libs/binaries are not fit for release

configure:
--enable/disable-libwebpmux; default is disabled.

makefile.unix:
src/mux/libwebpmux.a and examples/webpmux must be explicitly specified

Makefile.vc:
$(DIRLIB)\libwebpmux.lib and $(DIRBIN)\webpmux.exe must be explicitly
specified

Change-Id: I8246746b256010dd2a2e4de58291222d7eaf0457
2012-07-11 14:25:41 -07:00
James Zern
6f76d246ba cwebp: add basic TIFF support
Change-Id: I29f354e1ef9cf8f23002cf2e15814a53a94b4ed5
2012-07-02 11:22:52 -07:00
James Zern
fcec059322 makefile.unix: cwebp: fix OSX link
this is more a workaround than a fix. this change forces the tables
from yuv.h into the data section rather than generating them as common
blocks. prior to this the linkage would fail with e.g.,

Undefined symbols:
...
  "_VP8kClip4Bits", referenced from:
      _Yuv444ToRgba4444 in libwebp.a(libwebpdsp_la-upsampling.o)

broken since:
 48f8275 add colorspace for premultiplied alpha

Change-Id: I69c36758c31cd3a4b3ea5c412674d1a47b45447e
2012-06-26 18:09:55 -07:00
James Zern
e5f467420d add demux.c to the makefiles
Change-Id: If98c0ea859c845261218196df35e905589f09e76
2012-06-21 23:21:41 -07:00
Pascal Massimino
78f3e34504 Enable lossless encoder code
Remove USE_LOSSLESS_ENCODER compile flag
Update Makefile.am and makefile.unix

Change-Id: If7080c4d8f37994c7c784730c5e547bb0a851455
2012-06-13 00:26:58 -07:00
Urvang Joshi
c13f663261 Move consts to internal header format_constants.h
Change-Id: Ic6180c16d0b4245680738992925e4608c593fbe8
2012-05-24 15:02:02 +05:30
Pascal Massimino
3e863dda61 remove tcoder, switch alpha-plane compression to lossless
* Method #1 is now calling the lossless encoder on the alpha plane.
Format is not final, it's just a first draft. We need ad-hoc functions.
* removed now useless utils/alpha.*
* added utils/quant_levels.h instead
* removed the TCoder code altogether

Change-Id: I636840b6129a43171b74860e0a0fc5bb1bcffc6a
2012-05-21 06:24:48 -07:00
James Zern
00b29e282f normalize example header includes
headers from examples/ aren't installed so prefix with './' to be
explicit.

Change-Id: Ie6c1544a026a9859b7fc0cf238b21e031ef0013f
2012-05-15 13:48:11 -07:00
James Zern
061263a787 add example_util.[hc]
moves ReadFile to a common location

Change-Id: Ia81230671f16d7d4d218b9954a5be55577a85413
2012-05-15 13:42:57 -07:00
James Zern
4105061840 makefile.unix: add support for building vwebp
standalone from the normal examples due to the additional OpenGL
dependencies.
$ make -f makefile.unix examples/vwebp

Change-Id: I7e3f0b5e0328230cb32acdd1e2a011cbbb80e55d
2012-05-09 15:47:00 -07:00
James Zern
7885f8b25d makefile.unix: add lossless encoder files
comment out 'EXTRA_FLAGS += -DUSE_LOSSLESS_ENCODER' to disable

Change-Id: I75ead699c95644e319e7cbbf2a167d98e826fb7c
2012-05-08 11:25:28 -07:00
James Zern
d3adc81db2 makefile.unix add missing header files
Change-Id: I440ea805fd14884242bab2577e8286088616e5c7
2012-04-19 16:13:26 -07:00
James Zern
3f8ec1c21f makefile.unix & Android.mk: cosmetics
split object/source list to single line & sort
remove unnecessary tabs

Change-Id: I1db03555f9578e128c746853084438da2b9186eb
2012-04-12 12:29:57 -07:00
James Zern
f2623dbe58 enable lossless decoder
import changes from experimental 5529a2e^
and enable build in autoconf and makefile.unix; windows will be treated
separately.

Change-Id: Ie2e177a99db63190b4cd647b3edee3b4e13719e9
2012-04-10 23:06:36 -07:00
Pascal Massimino
61c2d51fd7 move the rescaling code into its own file and make enc/ and dec/ use it.
(cherry picked from commit 8e92d9e380a89b7443a2e2c3d16ce5a222e8c1e8)

Conflicts:

	Android.mk
	makefile.unix
	src/dec/vp8l.c
	src/utils/Makefile.am
2012-03-30 12:16:25 -07:00
James Zern
c274dc9679 makefile.unix: variable cosmetics
use CPPFLAGS to hold include paths, makes overriding CC a bit more
intuitive.

Change-Id: I6ecfc1957cae423dd150ebf37b878f1da6bbcef3
2012-01-25 10:46:28 -08:00
Pascal Massimino
b8375abd08 have makefile.unix clean up src/webp/*~ too
Change-Id: Ida17b92346055b197fe6ccb60a9ac223ddd46f90
2012-01-12 18:49:29 -08:00