Merge changes from libvpx/master by cherry-pick
This commit bring all up-to-date changes from master that are
applicable to nextgenv2. Due to the remove VP10 code in master,
we had to cherry pick the following commits to get those changes:
Add default flags for arm64/armv8 builds
Allows building simple targets with sane default flags.
For example, using the Android arm64 toolchain from the NDK:
https://developer.android.com/ndk/guides/standalone_toolchain.html
./build/tools/make-standalone-toolchain.sh --arch=arm64 \
--platform=android-24 --install-dir=/tmp/arm64
CROSS=/tmp/arm64/bin/aarch64-linux-android- \
~/libvpx/configure --target=arm64-linux-gcc --disable-multithread
BUG=webm:1143
vpx_lpf_horizontal_4_sse2: Remove dead load.
Change-Id: I51026c52baa1f0881fcd5b68e1fdf08a2dc0916e
Fail early when android target does not include --sdk-path
Change-Id: I07e7e63476a2e32e3aae123abdee8b7bbbdc6a8c
configure: clean up var style and set_all usage
Use quotes whenever possible and {} always for variables.
Replace multiple set_all calls with *able_feature().
Conflicts:
build/make/configure.sh
vp9-svc: Remove some unneeded code/comment.
datarate_test,DatarateTestLarge: normalize bits type
quiets a msvc warning:
conversion from 'const int64_t' to 'size_t', possible loss of data
mips added p6600 cpu support
Removed -funroll-loops
psnr.c: use int64_t for sum of differences
Since the values can be negative.
*.asm: normalize label format
add a trailing ':', though it's optional with the tools we support, it's
more common to use it to mark a label. this also quiets the
orphan-labels warning with nasm/yasm.
BUG=b/29583530
Prevent negative variance
Due to rounding, hbd variance may become negative. This commit put in
check and clamp of negative values to 0.
configure: remove old visual studio support (<2010)
BUG=b/29583530
Conflicts:
configure
configure: restore vs_version variable
inadvertently lost in the final patchset of:
078dff7 configure: remove old visual studio support (<2010)
this prevents an empty CONFIG_VS_VERSION and avoids make failure
Require x86inc.asm
Force enable x86inc.asm when building for x86. Previously there were
compatibility issues so a flag was added to simplify disabling this
code.
The known issues have been resolved and x86inc.asm is the preferred
abstraction layer (over x86_abi_support.asm).
BUG=b:29583530
convolve_test: fix byte offsets in hbd build
CONVERT_TO_BYTEPTR(x) was corrected in:
003a9d2 Port metric computation changes from nextgenv2
to use the more common (x) within the expansion. offsets should occur
after converting the pointer to the desired type.
+ factorized some common expressions
Conflicts:
test/convolve_test.cc
vpx_dsp: remove x86inc.asm distinction
BUG=b:29583530
Conflicts:
vpx_dsp/vpx_dsp.mk
vpx_dsp/vpx_dsp_rtcd_defs.pl
vpx_dsp/x86/highbd_variance_sse2.c
vpx_dsp/x86/variance_sse2.c
test: remove x86inc.asm distinction
BUG=b:29583530
Conflicts:
test/vp9_subtract_test.cc
configure: remove x86inc.asm distinction
BUG=b:29583530
Change-Id: I59a1192142e89a6a36b906f65a491a734e603617
Update vpx subpixel 1d filter ssse3 asm
Speed test shows the new vertical filters have degradation on Celeron
Chromebook. Added "X86_SUBPIX_VFILTER_PREFER_SLOW_CELERON" to control
the vertical filters activated code. Now just simply active the code
without degradation on Celeron. Later there should be 2 set of vertical
filters ssse3 functions, and let jump table to choose based on CPU type.
improve vpx_filter_block1d* based on replace paddsw+psrlw to pmulhrsw
Make set_reference control API work in VP9
Moved the API patch from NextGenv2. An example was included.
To try it, for example, run the following command:
$ examples/vpx_cx_set_ref vp9 352 288 in.yuv out.ivf 4 30
Conflicts:
examples.mk
examples/vpx_cx_set_ref.c
test/cx_set_ref.sh
vp9/decoder/vp9_decoder.c
deblock filter : moved from vp8 code branch
The deblocking filters used in vp8 have been moved to vpx_dsp for
use by both vp8 and vp9.
vpx_thread.[hc]: update webp source reference
+ drop the blob hash, the updated reference will be updated in the
commit message
BUG=b/29583578
vpx_thread: use native windows cond var if available
BUG=b/29583578
original webp change:
commit 110ad5835ecd66995d0e7f66dca1b90dea595f5a
Author: James Zern <jzern@google.com>
Date: Mon Nov 23 19:49:58 2015 -0800
thread: use native windows cond var if available
Vista / Server 2008 and up. no speed difference observed.
100644 blob 4fc372b7bc6980a9ed3618c8cce5b67ed7b0f412 src/utils/thread.c
100644 blob 840831185502d42a3246e4b7ff870121c8064791 src/utils/thread.h
vpx_thread: use InitializeCriticalSectionEx if available
BUG=b/29583578
original webp change:
commit 63fadc9ffacc77d4617526a50c696d21d558a70b
Author: James Zern <jzern@google.com>
Date: Mon Nov 23 20:38:46 2015 -0800
thread: use InitializeCriticalSectionEx if available
Windows Vista / Server 2008 and up
100644 blob f84207d89b3a6bb98bfe8f3fa55cad72dfd061ff src/utils/thread.c
100644 blob 840831185502d42a3246e4b7ff870121c8064791 src/utils/thread.h
vpx_thread: use WaitForSingleObjectEx if available
BUG=b/29583578
original webp change:
commit 0fd0e12bfe83f16ce4f1c038b251ccbc13c62ac2
Author: James Zern <jzern@google.com>
Date: Mon Nov 23 20:40:26 2015 -0800
thread: use WaitForSingleObjectEx if available
Windows XP and up
100644 blob d58f74e5523dbc985fc531cf5f0833f1e9157cf0 src/utils/thread.c
100644 blob 840831185502d42a3246e4b7ff870121c8064791 src/utils/thread.h
vpx_thread: use CreateThread for windows phone
BUG=b/29583578
original webp change:
commit d2afe974f9d751de144ef09d31255aea13b442c0
Author: James Zern <jzern@google.com>
Date: Mon Nov 23 20:41:26 2015 -0800
thread: use CreateThread for windows phone
_beginthreadex is unavailable for winrt/uwp
Change-Id: Ie7412a568278ac67f0047f1764e2521193d74d4d
100644 blob 93f7622797f05f6acc1126e8296c481d276e4047 src/utils/thread.c
100644 blob 840831185502d42a3246e4b7ff870121c8064791 src/utils/thread.h
vp9_postproc.c missing extern.
BUG=webm:1256
deblock: missing const on extern const.
postproc - move filling of noise buffer to vpx_dsp.
Fix encoder crashes for odd size input
clean-up vp9_intrapred_test
remove tuple and overkill VP9IntraPredBase class.
postproc: noise style fixes.
gtest-all.cc: quiet an unused variable warning
under windows / mingw builds
vp9_intrapred_test: follow-up cleanup
address few comments from ce050afaf3e288895c3bee4160336e2d2133b6ea
Change-Id: I3eece7efa9335f4210303993ef6c1857ad5c29c8
diff --git a/test/add_noise_test.cc b/test/add_noise_test.cc
index e9945c4..35aaadf 100644
--- a/test/add_noise_test.cc
+++ b/test/add_noise_test.cc
@@ -13,6 +13,7 @@
#include "third_party/googletest/src/include/gtest/gtest.h"
#include "./vpx_dsp_rtcd.h"
#include "vpx/vpx_integer.h"
+#include "vpx_dsp/postproc.h"
#include "vpx_mem/vpx_mem.h"
namespace {
@@ -40,50 +41,6 @@
return sqrt(v);
}
-// TODO(jimbankoski): The following 2 functions are duplicated in each codec.
-// For now the vp9 one has been copied into the test as is. We should normalize
-// these in vpx_dsp and not have 3 copies of these unless there is different
-// noise we add for each codec.
-
-double gaussian(double sigma, double mu, double x) {
- return 1 / (sigma * sqrt(2.0 * 3.14159265)) *
- (exp(-(x - mu) * (x - mu) / (2 * sigma * sigma)));
-}
-
-int setup_noise(int size_noise, char *noise) {
- char char_dist[300];
- const int ai = 4;
- const int qi = 24;
- const double sigma = ai + .5 + .6 * (63 - qi) / 63.0;
-
- /* set up a lookup table of 256 entries that matches
- * a gaussian distribution with sigma determined by q.
- */
- int next = 0;
-
- for (int i = -32; i < 32; i++) {
- int a_i = (int) (0.5 + 256 * gaussian(sigma, 0, i));
-
- if (a_i) {
- for (int j = 0; j < a_i; j++) {
- char_dist[next + j] = (char)(i);
- }
-
- next = next + a_i;
- }
- }
-
- for (; next < 256; next++)
- char_dist[next] = 0;
-
- for (int i = 0; i < size_noise; i++) {
- noise[i] = char_dist[rand() & 0xff]; // NOLINT
- }
-
- // Returns the most negative value in distribution.
- return char_dist[0];
-}
-
TEST_P(AddNoiseTest, CheckNoiseAdded) {
DECLARE_ALIGNED(16, char, blackclamp[16]);
DECLARE_ALIGNED(16, char, whiteclamp[16]);
@@ -92,12 +49,12 @@
const int height = 64;
const int image_size = width * height;
char noise[3072];
+ const int clamp = vpx_setup_noise(4.4, sizeof(noise), noise);
- const int clamp = setup_noise(3072, noise);
for (int i = 0; i < 16; i++) {
- blackclamp[i] = -clamp;
- whiteclamp[i] = -clamp;
- bothclamp[i] = -2 * clamp;
+ blackclamp[i] = clamp;
+ whiteclamp[i] = clamp;
+ bothclamp[i] = 2 * clamp;
}
uint8_t *const s = reinterpret_cast<uint8_t *>(vpx_calloc(image_size, 1));
@@ -127,7 +84,7 @@
// Check to make sure don't roll over.
for (int i = 0; i < image_size; ++i) {
- EXPECT_GT((int)s[i], 10) << "i = " << i;
+ EXPECT_GT(static_cast<int>(s[i]), clamp) << "i = " << i;
}
// Initialize pixels in the image to 0 and check for roll under.
@@ -138,7 +95,7 @@
// Check to make sure don't roll under.
for (int i = 0; i < image_size; ++i) {
- EXPECT_LT((int)s[i], 245) << "i = " << i;
+ EXPECT_LT(static_cast<int>(s[i]), 255 - clamp) << "i = " << i;
}
vpx_free(s);
@@ -153,11 +110,12 @@
const int image_size = width * height;
char noise[3072];
- const int clamp = setup_noise(3072, noise);
+ const int clamp = vpx_setup_noise(4.4, sizeof(noise), noise);
+
for (int i = 0; i < 16; i++) {
- blackclamp[i] = -clamp;
- whiteclamp[i] = -clamp;
- bothclamp[i] = -2 * clamp;
+ blackclamp[i] = clamp;
+ whiteclamp[i] = clamp;
+ bothclamp[i] = 2 * clamp;
}
uint8_t *const s = reinterpret_cast<uint8_t *>(vpx_calloc(image_size, 1));
@@ -175,7 +133,7 @@
width, height, width));
for (int i = 0; i < image_size; ++i) {
- EXPECT_EQ((int)s[i], (int)d[i]) << "i = " << i;
+ EXPECT_EQ(static_cast<int>(s[i]), static_cast<int>(d[i])) << "i = " << i;
}
vpx_free(d);
