the 'accum' variable can be larger than 15b for large
rescale values.
Assert triggered:
src/dsp/rescaler_sse2.c:249: RescalerExportRowExpand_SSE2: Assertion `v >= 0 && v <= 255' failed.
src/dsp/rescaler_sse2.c:350: RescalerExportRowShrink_SSE2: Assertion `v >= 0 && v <= 255' failed.
-> fall back to C implementation in this case for now
Change-Id: I7ea1cb72301cafc1459be403f6a6f4e3cbc89bb1
with WEBP_NEON_OMIT_C_CODE the default _C functions won't be set and
with WEBP_REDUCE_CSP the NEON functions won't be either triggering an
assert for an empty table member.
BUG=chromium:792627
Change-Id: I8d2d430eaa37bb92885b61a3dd39f961924a8def
RGB to YUV conversion was not using SSE to finish up the row.
End data is now copied to a buffer big enough to fit in a
SSE register.
(UPSAMPLE_LAST_BLOCK was already using that trick).
Change-Id: Ie539bcbe570a643a774aa88263503c0d2c41890f
only supported ones are: RGBA/BGRA/rgbA/bgrA (decoder)
as well as: WebPPictureImportRGB/RGBX/RGBA (encoder).
(note: extras/get_disto is affected too)
Change-Id: If6c4f95054ca15759c4e289fb3b4c352b3521c2c
remove auto-filter (-af) support and make WebPPictureCopy,
WebPPictureIsView, WebPPictureView, WebPPictureCrop, and
WebPPictureRescale noops.
Change-Id: If39d512cc268a0015298a1138dbc94feb86575e5
with gcc-4.8, clang-4.0.1/5 this is no faster (actually up to 2x slower)
than the code generated for memset (0x01010... * dst[-1]). shuffles in
sse4 recover a bit, but performance is still down.
Change-Id: Ie85e8353f8ede559d0b05a1d388787fd18ecc80f
Rewrote WebPPictureHasTransparency() to use them (even for argb).
This is 10% faster, for some reasons.
SSE2 version should be straightforward.
Removes a TODO.
Change-Id: I7ad5848fc5e355e2df505dbcd5a0f42fb6cbab41
when targeting NEON C functions with NEON equivalents won't be used, but
will contribute to binary size. the same goes for sse2, etc., but this
change is primarily concerned with binary sizes for android arm targets.
note '-noasm' or otherwise modifying VP8GetCPUInfo will have no effect
on the use of NEON functions.
this decision can be overridden by defining WEBP_DSP_OMIT_C_CODE to 0.
Change-Id: I47bd453c84a3d341ca39bc986a39eb9c785aface
and all older versions.
force Sub3() to not be inlined, otherwise the code in Select() will be
incorrect.
extends the check add previously in:
637b3888 dsp/lossless: workaround gcc-4.9 bug on arm
BUG=webp:363
Change-Id: I1403b558f8660b764f3a570a3326822d5ef0be29
ssse3 is bit #9 in ecx, bit 1 is sse3. this only controls the check for
slow ssse3 and likely had no ill effect.
Change-Id: I84ce73dc480e1cdbd085e37be06f3f402116c201
add a check for cpu-features.h and rework some of the ifdef's around
android + neon. for android builds with cpu-features enabled the
*_neon.c files will still need to be flagged correctly (with e.g.,
.c.neon in Android.mk) to properly build them.
BUG=webp:353
Change-Id: I905ce305af0a204e560b915d8665093a3edaceb9
including the type in the macro doesn't bring much benefit to ordering,
current platforms work with a prefix, this would be insufficient if the
attribute needed to follow the function prototype. this form makes it
easier to override on the command line.
BUG=webp:355
Change-Id: Iba41ec0bb319403054be0e899c4cc472dd932fd9
Half of the functionality was duplicated.
The rest is about the alpha channel handling so we
might as well put it in the appropriate file.
Change-Id: I8d5ef0afce82cc4842ab7132fd97995c42e6140a
Currently, none are available. If WEBP_HAVE_SSE2 eventually works,
we'll have to refine this conditionals.
BUG=webp:261
Change-Id: Ibc63ee1c013f2a4169eeb85cc8b6317b6420c2ad
this avoids duplicates between these trees and dsp/, e.g., enc/tree.c,
dec/tree.c, making pulling the whole library source tree into one target
possible
BUG=webp:279
Change-Id: I060a614833c7c24ddd37bf641702ae6a5eef1775
vmlal_u8() is prone to overflow during the accumulation.
There was a mismatch happening at low q mostly. Because in this
case the distortion is important and the accumulated sum was
later than 16bit-unsigned.
Change-Id: I1a08a2f744bcdf0b26647e61b9ee92a0c2e28fe8
This is meant to be used for run-time detection of slow platforms
regarding instructions like pshufb and bsr.
Adapted from libvpx patch: https://chromium-review.googlesource.com/#/c/367731
Change-Id: I2c22fbb9aae699d87a041393ba1ad5f1f21ff640
and 15% faster MultARGBRow()
by switching to formulae:
X / 255 = (X + 1 + (X >> 8)) >> 8 for any 16bit value X.
(X / 255 + .5) = (XX + (XX >> 8)) >> 8, with XX = X + 128
Change-Id: Ia4a7408aee74d7f61b58f5dff304d05546c04e81
after:
fbba5bc optimize predictor #1 in plain-C For some reason, gcc has hard
time inlining this one...
Change-Id: I2e2416593acd4c9d14958d8757bfd284d999100b
For some reason, gcc has hard time inlining this one...
Also optimize predictor #0 and #1 for encoding, so we don't have to
call the generic pointers VP8LPredictors[...]
Change-Id: I1ff31e3b83874b53f84fe23487f644619fd61db9
Average3 created a slowdown of 1-2% in lossless decoding.
Average4 created a slowdown of 2-3% in lossless decoding.
Change-Id: Ic2e62cdd83fc897887ec2bf41ea7cadbada84fe5
...instead of the pointers stored in the array.
Should be faster (inlined) and safer.
Also: suffix explicitly the functions with _SSE2
Change-Id: Ie7de4b8876caea15067fdbe44abfedd72b299a90
Before, a first thread could enter VP8LDspInitSSE2, set
VP8LPredictorsAdd to an SSE2 version BEFORE another thread
would do the memcpy from VP8LPredictorsAdd to VP8LPredictorsAdd_C
thus leading to a C version actually being the SSE2 one (which
would then create an infinite recursion in the SSE2 predictors
at execution).
Change-Id: I224f4ceab31d38f77a1375a7e2636a6014080e3a
Benchmarks from vrabaud@:
8BIT/GRAY corpus speed: faster: -4.3 % , corpus size: unchanged
skal/sources_png_skal corpus speed: faster: -5.2 % , corpus size: unchanged
images/png_rgb corpus speed: faster: -5.1 % , corpus size: unchanged
images/lpcb corpus speed: unchanged, corpus size: unchanged
images/png_big corpus speed: faster: -1.7 % , corpus size: unchanged
images/png_doc corpus speed: unchanged, corpus size: unchanged
images/png_1bit corpus speed: faster: -1.2 % , corpus size: unchanged
images/jpeg_small corpus speed: unchanged, corpus size: unchanged
images/icip_core1 corpus speed: unchanged, corpus size: unchanged
images/png_gray corpus speed: faster: -2.5 % , corpus size: unchanged
images/jpeg_high_quality corpus speed: faster: -4.0 % , corpus size: unchanged
images/jpeg corpus speed: faster: -2.3 % , corpus size: unchanged
images/png_translucent corpus speed: faster: -2.8 % , corpus size: unchanged
images/gif corpus speed: faster: -1.4 % , corpus size: unchanged
images/png_opaque corpus speed: faster: -2.8 % , corpus size: unchanged
images/png_rgb_opaque corpus speed: unchanged, corpus size: unchanged
images/png_indexed corpus speed: faster: -2.0 % , corpus size: unchanged
images/all corpus speed: faster: -1.5 % , corpus size: unchanged
images/png_small corpus speed: unchanged, corpus size: unchanged
images/png corpus speed: unchanged, corpus size: unchanged
images/gif_still corpus speed: faster: -1.6 % , corpus size: unchanged
Change-Id: I69fe11baa188c5d32cbc77a84b8c0deae13d792b
Roughly, if both the source and the reference areas are
darker too dark (R/G/B <= ~6), they are ignored.
One caveat: SSIM calculation won't work for U/V planes,
which are 128-centered and not related to luminance.
But WebPPlaneDistortion() enforces the conversion to RGB,
if needed.
Change-Id: I586c2579c475583b8c90c5baefd766b1d5aea591
* prevent 64bit overflow by controlling the 32b->64b conversions
and preventively descaling by 8bit before the final multiply
* adjust the threshold constants C1 and C2 to de-emphasis the dark
areas
* use a hat-like filter instead of box-filtering to avoid blockiness
during averaging
SSIM distortion calc is actually *faster* now in SSE2, because of the
unrolling during the function rewrite.
The C-version is quite slower because still un-optimized.
Change-Id: I96e2715827f79d26faae354cc28c7406c6800c90
