summaryrefslogtreecommitdiffstats
path: root/media/libvpx/vp8/common/x86/vp8_variance_sse2.c
diff options
context:
space:
mode:
authorMatt A. Tobin <mattatobin@localhost.localdomain>2018-02-02 04:16:08 -0500
committerMatt A. Tobin <mattatobin@localhost.localdomain>2018-02-02 04:16:08 -0500
commit5f8de423f190bbb79a62f804151bc24824fa32d8 (patch)
tree10027f336435511475e392454359edea8e25895d /media/libvpx/vp8/common/x86/vp8_variance_sse2.c
parent49ee0794b5d912db1f95dce6eb52d781dc210db5 (diff)
downloadUXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.gz
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.lz
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.xz
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.zip
Add m-esr52 at 52.6.0
Diffstat (limited to 'media/libvpx/vp8/common/x86/vp8_variance_sse2.c')
-rw-r--r--media/libvpx/vp8/common/x86/vp8_variance_sse2.c403
1 files changed, 403 insertions, 0 deletions
diff --git a/media/libvpx/vp8/common/x86/vp8_variance_sse2.c b/media/libvpx/vp8/common/x86/vp8_variance_sse2.c
new file mode 100644
index 000000000..1c15ed880
--- /dev/null
+++ b/media/libvpx/vp8/common/x86/vp8_variance_sse2.c
@@ -0,0 +1,403 @@
+/*
+ * Copyright (c) 2010 The WebM project authors. All Rights Reserved.
+ *
+ * Use of this source code is governed by a BSD-style license
+ * that can be found in the LICENSE file in the root of the source
+ * tree. An additional intellectual property rights grant can be found
+ * in the file PATENTS. All contributing project authors may
+ * be found in the AUTHORS file in the root of the source tree.
+ */
+
+#include "./vp8_rtcd.h"
+#include "vpx_config.h"
+#include "vp8/common/variance.h"
+#include "vpx_ports/mem.h"
+#include "vp8/common/x86/filter_x86.h"
+
+extern void filter_block1d_h6_mmx(const unsigned char *src_ptr, unsigned short *output_ptr, unsigned int src_pixels_per_line, unsigned int pixel_step, unsigned int output_height, unsigned int output_width, short *filter);
+extern void filter_block1d_v6_mmx(const short *src_ptr, unsigned char *output_ptr, unsigned int pixels_per_line, unsigned int pixel_step, unsigned int output_height, unsigned int output_width, short *filter);
+extern void filter_block1d8_h6_sse2(const unsigned char *src_ptr, unsigned short *output_ptr, unsigned int src_pixels_per_line, unsigned int pixel_step, unsigned int output_height, unsigned int output_width, short *filter);
+extern void filter_block1d8_v6_sse2(const short *src_ptr, unsigned char *output_ptr, unsigned int pixels_per_line, unsigned int pixel_step, unsigned int output_height, unsigned int output_width, short *filter);
+
+extern void vp8_filter_block2d_bil4x4_var_mmx
+(
+ const unsigned char *ref_ptr,
+ int ref_pixels_per_line,
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ const short *HFilter,
+ const short *VFilter,
+ int *sum,
+ unsigned int *sumsquared
+);
+
+void vp8_filter_block2d_bil_var_sse2
+(
+ const unsigned char *ref_ptr,
+ int ref_pixels_per_line,
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ unsigned int Height,
+ int xoffset,
+ int yoffset,
+ int *sum,
+ unsigned int *sumsquared
+);
+void vp8_half_horiz_vert_variance8x_h_sse2
+(
+ const unsigned char *ref_ptr,
+ int ref_pixels_per_line,
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ unsigned int Height,
+ int *sum,
+ unsigned int *sumsquared
+);
+void vp8_half_horiz_vert_variance16x_h_sse2
+(
+ const unsigned char *ref_ptr,
+ int ref_pixels_per_line,
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ unsigned int Height,
+ int *sum,
+ unsigned int *sumsquared
+);
+void vp8_half_horiz_variance8x_h_sse2
+(
+ const unsigned char *ref_ptr,
+ int ref_pixels_per_line,
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ unsigned int Height,
+ int *sum,
+ unsigned int *sumsquared
+);
+void vp8_half_horiz_variance16x_h_sse2
+(
+ const unsigned char *ref_ptr,
+ int ref_pixels_per_line,
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ unsigned int Height,
+ int *sum,
+ unsigned int *sumsquared
+);
+void vp8_half_vert_variance8x_h_sse2
+(
+ const unsigned char *ref_ptr,
+ int ref_pixels_per_line,
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ unsigned int Height,
+ int *sum,
+ unsigned int *sumsquared
+);
+void vp8_half_vert_variance16x_h_sse2
+(
+ const unsigned char *ref_ptr,
+ int ref_pixels_per_line,
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ unsigned int Height,
+ int *sum,
+ unsigned int *sumsquared
+);
+
+unsigned int vp8_sub_pixel_variance4x4_wmt
+(
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ int xoffset,
+ int yoffset,
+ const unsigned char *dst_ptr,
+ int dst_pixels_per_line,
+ unsigned int *sse
+)
+{
+ int xsum;
+ unsigned int xxsum;
+ vp8_filter_block2d_bil4x4_var_mmx(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line,
+ vp8_bilinear_filters_x86_4[xoffset], vp8_bilinear_filters_x86_4[yoffset],
+ &xsum, &xxsum
+ );
+ *sse = xxsum;
+ return (xxsum - (((unsigned int)xsum * xsum) >> 4));
+}
+
+
+unsigned int vp8_sub_pixel_variance8x8_wmt
+(
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ int xoffset,
+ int yoffset,
+ const unsigned char *dst_ptr,
+ int dst_pixels_per_line,
+ unsigned int *sse
+)
+{
+ int xsum;
+ unsigned int xxsum;
+
+ if (xoffset == 4 && yoffset == 0)
+ {
+ vp8_half_horiz_variance8x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 8,
+ &xsum, &xxsum);
+ }
+ else if (xoffset == 0 && yoffset == 4)
+ {
+ vp8_half_vert_variance8x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 8,
+ &xsum, &xxsum);
+ }
+ else if (xoffset == 4 && yoffset == 4)
+ {
+ vp8_half_horiz_vert_variance8x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 8,
+ &xsum, &xxsum);
+ }
+ else
+ {
+ vp8_filter_block2d_bil_var_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 8,
+ xoffset, yoffset,
+ &xsum, &xxsum);
+ }
+
+ *sse = xxsum;
+ return (xxsum - (((unsigned int)xsum * xsum) >> 6));
+}
+
+unsigned int vp8_sub_pixel_variance16x16_wmt
+(
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ int xoffset,
+ int yoffset,
+ const unsigned char *dst_ptr,
+ int dst_pixels_per_line,
+ unsigned int *sse
+)
+{
+ int xsum0, xsum1;
+ unsigned int xxsum0, xxsum1;
+
+
+ /* note we could avoid these if statements if the calling function
+ * just called the appropriate functions inside.
+ */
+ if (xoffset == 4 && yoffset == 0)
+ {
+ vp8_half_horiz_variance16x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ &xsum0, &xxsum0);
+ }
+ else if (xoffset == 0 && yoffset == 4)
+ {
+ vp8_half_vert_variance16x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ &xsum0, &xxsum0);
+ }
+ else if (xoffset == 4 && yoffset == 4)
+ {
+ vp8_half_horiz_vert_variance16x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ &xsum0, &xxsum0);
+ }
+ else
+ {
+ vp8_filter_block2d_bil_var_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ xoffset, yoffset,
+ &xsum0, &xxsum0
+ );
+
+ vp8_filter_block2d_bil_var_sse2(
+ src_ptr + 8, src_pixels_per_line,
+ dst_ptr + 8, dst_pixels_per_line, 16,
+ xoffset, yoffset,
+ &xsum1, &xxsum1
+ );
+ xsum0 += xsum1;
+ xxsum0 += xxsum1;
+ }
+
+ *sse = xxsum0;
+ return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 8));
+}
+
+unsigned int vp8_sub_pixel_variance16x8_wmt
+(
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ int xoffset,
+ int yoffset,
+ const unsigned char *dst_ptr,
+ int dst_pixels_per_line,
+ unsigned int *sse
+
+)
+{
+ int xsum0, xsum1;
+ unsigned int xxsum0, xxsum1;
+
+ if (xoffset == 4 && yoffset == 0)
+ {
+ vp8_half_horiz_variance16x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 8,
+ &xsum0, &xxsum0);
+ }
+ else if (xoffset == 0 && yoffset == 4)
+ {
+ vp8_half_vert_variance16x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 8,
+ &xsum0, &xxsum0);
+ }
+ else if (xoffset == 4 && yoffset == 4)
+ {
+ vp8_half_horiz_vert_variance16x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 8,
+ &xsum0, &xxsum0);
+ }
+ else
+ {
+ vp8_filter_block2d_bil_var_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 8,
+ xoffset, yoffset,
+ &xsum0, &xxsum0);
+
+ vp8_filter_block2d_bil_var_sse2(
+ src_ptr + 8, src_pixels_per_line,
+ dst_ptr + 8, dst_pixels_per_line, 8,
+ xoffset, yoffset,
+ &xsum1, &xxsum1);
+ xsum0 += xsum1;
+ xxsum0 += xxsum1;
+ }
+
+ *sse = xxsum0;
+ return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 7));
+}
+
+unsigned int vp8_sub_pixel_variance8x16_wmt
+(
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ int xoffset,
+ int yoffset,
+ const unsigned char *dst_ptr,
+ int dst_pixels_per_line,
+ unsigned int *sse
+)
+{
+ int xsum;
+ unsigned int xxsum;
+
+ if (xoffset == 4 && yoffset == 0)
+ {
+ vp8_half_horiz_variance8x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ &xsum, &xxsum);
+ }
+ else if (xoffset == 0 && yoffset == 4)
+ {
+ vp8_half_vert_variance8x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ &xsum, &xxsum);
+ }
+ else if (xoffset == 4 && yoffset == 4)
+ {
+ vp8_half_horiz_vert_variance8x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ &xsum, &xxsum);
+ }
+ else
+ {
+ vp8_filter_block2d_bil_var_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ xoffset, yoffset,
+ &xsum, &xxsum);
+ }
+
+ *sse = xxsum;
+ return (xxsum - (((unsigned int)xsum * xsum) >> 7));
+}
+
+
+unsigned int vp8_variance_halfpixvar16x16_h_wmt(
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ const unsigned char *dst_ptr,
+ int dst_pixels_per_line,
+ unsigned int *sse)
+{
+ int xsum0;
+ unsigned int xxsum0;
+
+ vp8_half_horiz_variance16x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ &xsum0, &xxsum0);
+
+ *sse = xxsum0;
+ return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 8));
+}
+
+
+unsigned int vp8_variance_halfpixvar16x16_v_wmt(
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ const unsigned char *dst_ptr,
+ int dst_pixels_per_line,
+ unsigned int *sse)
+{
+ int xsum0;
+ unsigned int xxsum0;
+ vp8_half_vert_variance16x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ &xsum0, &xxsum0);
+
+ *sse = xxsum0;
+ return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 8));
+}
+
+
+unsigned int vp8_variance_halfpixvar16x16_hv_wmt(
+ const unsigned char *src_ptr,
+ int src_pixels_per_line,
+ const unsigned char *dst_ptr,
+ int dst_pixels_per_line,
+ unsigned int *sse)
+{
+ int xsum0;
+ unsigned int xxsum0;
+
+ vp8_half_horiz_vert_variance16x_h_sse2(
+ src_ptr, src_pixels_per_line,
+ dst_ptr, dst_pixels_per_line, 16,
+ &xsum0, &xxsum0);
+
+ *sse = xxsum0;
+ return (xxsum0 - (((unsigned int)xsum0 * xsum0) >> 8));
+}