Merge "removed reference to "LLM" and "x8"" into experimental

12 files changed, 54 insertions, 68 deletions

diff --git a/vp9/common/ppc/vp9_idctllm_altivec.asm b/vp9/common/ppc/vp9_idct_altivec.asm
index 117d9cfc8..b87aa4200 100644
--- a/vp9/common/ppc/vp9_idctllm_altivec.asm
+++ b/vp9/common/ppc/vp9_idct_altivec.asm
@@ -9,7 +9,7 @@
 ;
 
 
-    .globl short_idct4x4llm_ppc
+    .globl short_idct4x4_ppc
 
 .macro load_c V, LABEL, OFF, R0, R1
     lis     \R0, \LABEL@ha
@@ -21,7 +21,7 @@
 ;# r4 short *output
 ;# r5 int pitch
     .align 2
-short_idct4x4llm_ppc:
+short_idct4x4_ppc:
     mfspr   r11, 256            ;# get old VRSAVE
     oris    r12, r11, 0xfff8
     mtspr   256, r12            ;# set VRSAVE
diff --git a/vp9/common/ppc/vp9_systemdependent.c b/vp9/common/ppc/vp9_systemdependent.c
index 02035191f..ac13722d4 100644
--- a/vp9/common/ppc/vp9_systemdependent.c
+++ b/vp9/common/ppc/vp9_systemdependent.c
@@ -63,7 +63,7 @@ void recon_b_ppc(short *diff_ptr, unsigned char *pred_ptr, unsigned char *dst_pt
 void recon2b_ppc(short *diff_ptr, unsigned char *pred_ptr, unsigned char *dst_ptr, int stride);
 void recon4b_ppc(short *diff_ptr, unsigned char *pred_ptr, unsigned char *dst_ptr, int stride);
 
-extern void short_idct4x4llm_ppc(short *input, short *output, int pitch);
+extern void short_idct4x4_ppc(short *input, short *output, int pitch);
 
 // Generic C
 extern subpixel_predict_function vp9_sixtap_predict_c;
@@ -83,8 +83,8 @@ void vp9_recon_b_c(short *diff_ptr, unsigned char *pred_ptr, unsigned char *dst_
 void vp9_recon2b_c(short *diff_ptr, unsigned char *pred_ptr, unsigned char *dst_ptr, int stride);
 void vp9_recon4b_c(short *diff_ptr, unsigned char *pred_ptr, unsigned char *dst_ptr, int stride);
 
-extern void vp9_short_idct4x4llm_1_c(short *input, short *output, int pitch);
-extern void vp9_short_idct4x4llm_c(short *input, short *output, int pitch);
+extern void vp9_short_idct4x4_1_c(short *input, short *output, int pitch);
+extern void vp9_short_idct4x4_c(short *input, short *output, int pitch);
 extern void vp8_dc_only_idct_c(short input_dc, short *output, int pitch);
 
 // PPC
@@ -139,8 +139,8 @@ void vp9_machine_specific_config(void) {
   vp9_sixtap_predict8x4                = sixtap_predict8x4_ppc;
   vp9_sixtap_predict                   = sixtap_predict_ppc;
 
-  vp8_short_idct4x4_1                  = vp9_short_idct4x4llm_1_c;
-  vp8_short_idct4x4                    = short_idct4x4llm_ppc;
+  vp8_short_idct4x4_1                  = vp9_short_idct4x4_1_c;
+  vp8_short_idct4x4                    = short_idct4x4_ppc;
   vp8_dc_only_idct                      = vp8_dc_only_idct_c;
 
   vp8_lf_mbvfull                       = loop_filter_mbv_ppc;
diff --git a/vp9/common/vp9_idctllm.c b/vp9/common/vp9_idct.c
index e2106250f..3ec093f73 100644
--- a/vp9/common/vp9_idctllm.c
+++ b/vp9/common/vp9_idct.c
@@ -8,20 +8,6 @@
  *  be found in the AUTHORS file in the root of the source tree.
  */
 
-
-/****************************************************************************
- * Notes:
- *
- * This implementation makes use of 16 bit fixed point verio of two multiply
- * constants:
- *         1.   sqrt(2) * cos (pi/8)
- *         2.   sqrt(2) * sin (pi/8)
- * Becuase the first constant is bigger than 1, to maintain the same 16 bit
- * fixed point precision as the second one, we use a trick of
- *         x * a = x + x*(a-1)
- * so
- *         x * sqrt(2) * cos (pi/8) = x + x * (sqrt(2) *cos(pi/8)-1).
- **************************************************************************/
 #include <assert.h>
 #include <math.h>
 
@@ -32,7 +18,7 @@
 #include "vp9/common/vp9_common.h"
 #include "vp9/common/vp9_idct.h"
 
-void vp9_short_inv_walsh4x4_x8_c(int16_t *input, int16_t *output, int pitch) {
+void vp9_short_iwalsh4x4_c(int16_t *input, int16_t *output, int pitch) {
   int i;
   int a1, b1, c1, d1;
   int16_t *ip = input;
@@ -73,7 +59,7 @@ void vp9_short_inv_walsh4x4_x8_c(int16_t *input, int16_t *output, int pitch) {
   }
 }
 
-void vp9_short_inv_walsh4x4_1_x8_c(int16_t *in, int16_t *out, int pitch) {
+void vp9_short_iwalsh4x4_1_c(int16_t *in, int16_t *out, int pitch) {
   int i;
   int16_t tmp[4];
   int16_t *ip = in;
@@ -99,7 +85,7 @@ void vp9_dc_only_inv_walsh_add_c(int input_dc, uint8_t *pred_ptr,
   int r, c;
   int16_t dc = input_dc;
   int16_t tmp[4 * 4];
-  vp9_short_inv_walsh4x4_1_x8_c(&dc, tmp, 4 << 1);
+  vp9_short_iwalsh4x4_1_c(&dc, tmp, 4 << 1);
 
   for (r = 0; r < 4; r++) {
     for (c = 0; c < 4; c++)
@@ -130,7 +116,7 @@ void vp9_idct4_1d_c(int16_t *input, int16_t *output) {
   output[3] = step[0] - step[3];
 }
 
-void vp9_short_idct4x4llm_c(int16_t *input, int16_t *output, int pitch) {
+void vp9_short_idct4x4_c(int16_t *input, int16_t *output, int pitch) {
   int16_t out[4 * 4];
   int16_t *outptr = out;
   const int half_pitch = pitch >> 1;
@@ -156,7 +142,7 @@ void vp9_short_idct4x4llm_c(int16_t *input, int16_t *output, int pitch) {
   }
 }
 
-void vp9_short_idct4x4llm_1_c(int16_t *input, int16_t *output, int pitch) {
+void vp9_short_idct4x4_1_c(int16_t *input, int16_t *output, int pitch) {
   int i;
   int a1;
   int16_t *op = output;
diff --git a/vp9/common/vp9_rtcd_defs.sh b/vp9/common/vp9_rtcd_defs.sh
index 911fcc55e..0c2a5c94a 100644
--- a/vp9/common/vp9_rtcd_defs.sh
+++ b/vp9/common/vp9_rtcd_defs.sh
@@ -279,11 +279,11 @@ specialize vp9_convolve8_avg_vert ssse3
 #
 # dct
 #
-prototype void vp9_short_idct4x4llm_1 "int16_t *input, int16_t *output, int pitch"
-specialize vp9_short_idct4x4llm_1
+prototype void vp9_short_idct4x4_1 "int16_t *input, int16_t *output, int pitch"
+specialize vp9_short_idct4x4_1
 
-prototype void vp9_short_idct4x4llm "int16_t *input, int16_t *output, int pitch"
-specialize vp9_short_idct4x4llm sse2
+prototype void vp9_short_idct4x4 "int16_t *input, int16_t *output, int pitch"
+specialize vp9_short_idct4x4 sse2
 
 prototype void vp9_short_idct8x8 "int16_t *input, int16_t *output, int pitch"
 specialize vp9_short_idct8x8
@@ -330,10 +330,10 @@ specialize vp9_idct4_1d sse2
 prototype void vp9_dc_only_idct_add "int input_dc, uint8_t *pred_ptr, uint8_t *dst_ptr, int pitch, int stride"
 specialize vp9_dc_only_idct_add sse2
 
-prototype void vp9_short_inv_walsh4x4_1_x8 "int16_t *input, int16_t *output, int pitch"
-specialize vp9_short_inv_walsh4x4_1_x8
-prototype void vp9_short_inv_walsh4x4_x8 "int16_t *input, int16_t *output, int pitch"
-specialize vp9_short_inv_walsh4x4_x8
+prototype void vp9_short_iwalsh4x4_1 "int16_t *input, int16_t *output, int pitch"
+specialize vp9_short_iwalsh4x4_1
+prototype void vp9_short_iwalsh4x4 "int16_t *input, int16_t *output, int pitch"
+specialize vp9_short_iwalsh4x4
 prototype void vp9_dc_only_inv_walsh_add "int input_dc, uint8_t *pred_ptr, uint8_t *dst_ptr, int pitch, int stride"
 specialize vp9_dc_only_inv_walsh_add
 
@@ -600,11 +600,11 @@ specialize vp9_short_fdct32x32
 prototype void vp9_short_fdct16x16 "int16_t *InputData, int16_t *OutputData, int pitch"
 specialize vp9_short_fdct16x16 sse2
 
-prototype void vp9_short_walsh4x4_x8 "int16_t *InputData, int16_t *OutputData, int pitch"
-specialize vp9_short_walsh4x4_x8
+prototype void vp9_short_walsh4x4 "int16_t *InputData, int16_t *OutputData, int pitch"
+specialize vp9_short_walsh4x4
 
-prototype void vp9_short_walsh8x4_x8 "int16_t *InputData, int16_t *OutputData, int pitch"
-specialize vp9_short_walsh8x4_x8
+prototype void vp9_short_walsh8x4 "int16_t *InputData, int16_t *OutputData, int pitch"
+specialize vp9_short_walsh8x4
 
 #
 # Motion search
diff --git a/vp9/common/x86/vp9_idctllm_sse2.asm b/vp9/common/x86/vp9_idct_sse2.asm
index 8f3c6dfc3..8f3c6dfc3 100644
--- a/vp9/common/x86/vp9_idctllm_sse2.asm
+++ b/vp9/common/x86/vp9_idct_sse2.asm
diff --git a/vp9/common/x86/vp9_idctllm_x86.c b/vp9/common/x86/vp9_idct_x86.c
index 3d7a1481c..6a35823bd 100644
--- a/vp9/common/x86/vp9_idctllm_x86.c
+++ b/vp9/common/x86/vp9_idct_x86.c
@@ -74,7 +74,7 @@ void vp9_dc_only_idct_add_sse2(int input_dc, uint8_t *pred_ptr,
   *(int *)dst_ptr = _mm_cvtsi128_si32(p1);
 }
 
-void vp9_short_idct4x4llm_sse2(int16_t *input, int16_t *output, int pitch) {
+void vp9_short_idct4x4_sse2(int16_t *input, int16_t *output, int pitch) {
   const __m128i zero = _mm_setzero_si128();
   const __m128i eight = _mm_set1_epi16(8);
   const __m128i cst = _mm_setr_epi16((int16_t)cospi_16_64, (int16_t)cospi_16_64,
diff --git a/vp9/decoder/vp9_decodframe.c b/vp9/decoder/vp9_decodframe.c
index 18f276e83..d1d07c753 100644
--- a/vp9/decoder/vp9_decodframe.c
+++ b/vp9/decoder/vp9_decodframe.c
@@ -138,14 +138,14 @@ static void mb_init_dequantizer(VP9D_COMP *pbi, MACROBLOCKD *mb) {
 
   if (mb->lossless) {
     assert(qindex == 0);
-    mb->inv_txm4x4_1      = vp9_short_inv_walsh4x4_1_x8;
-    mb->inv_txm4x4        = vp9_short_inv_walsh4x4_x8;
+    mb->inv_txm4x4_1      = vp9_short_iwalsh4x4_1;
+    mb->inv_txm4x4        = vp9_short_iwalsh4x4;
     mb->itxm_add          = vp9_dequant_idct_add_lossless_c;
     mb->itxm_add_y_block  = vp9_dequant_idct_add_y_block_lossless_c;
     mb->itxm_add_uv_block = vp9_dequant_idct_add_uv_block_lossless_c;
   } else {
-    mb->inv_txm4x4_1      = vp9_short_idct4x4llm_1;
-    mb->inv_txm4x4        = vp9_short_idct4x4llm;
+    mb->inv_txm4x4_1      = vp9_short_idct4x4_1;
+    mb->inv_txm4x4        = vp9_short_idct4x4;
     mb->itxm_add          = vp9_dequant_idct_add;
     mb->itxm_add_y_block  = vp9_dequant_idct_add_y_block;
     mb->itxm_add_uv_block = vp9_dequant_idct_add_uv_block;
diff --git a/vp9/decoder/vp9_dequantize.c b/vp9/decoder/vp9_dequantize.c
index 92b78ed19..cb4601a15 100644
--- a/vp9/decoder/vp9_dequantize.c
+++ b/vp9/decoder/vp9_dequantize.c
@@ -126,7 +126,7 @@ void vp9_dequant_idct_add_c(int16_t *input, const int16_t *dq, uint8_t *pred,
       input[i] *= dq[i];
 
     // the idct halves ( >> 1) the pitch
-    vp9_short_idct4x4llm(input, output, 4 << 1);
+    vp9_short_idct4x4(input, output, 4 << 1);
 
     vpx_memset(input, 0, 32);
 
@@ -148,7 +148,7 @@ void vp9_dequant_dc_idct_add_c(int16_t *input, const int16_t *dq, uint8_t *pred,
     input[i] *= dq[i];
 
   // the idct halves ( >> 1) the pitch
-  vp9_short_idct4x4llm(input, output, 4 << 1);
+  vp9_short_idct4x4(input, output, 4 << 1);
   vpx_memset(input, 0, 32);
   vp9_add_residual_4x4(output, pred, pitch, dest, stride);
 }
@@ -163,7 +163,7 @@ void vp9_dequant_idct_add_lossless_c(int16_t *input, const int16_t *dq,
     for (i = 0; i < 16; i++)
       input[i] *= dq[i];
 
-    vp9_short_inv_walsh4x4_x8_c(input, output, 4 << 1);
+    vp9_short_iwalsh4x4_c(input, output, 4 << 1);
 
     vpx_memset(input, 0, 32);
 
@@ -186,7 +186,7 @@ void vp9_dequant_dc_idct_add_lossless_c(int16_t *input, const int16_t *dq,
   for (i = 1; i < 16; i++)
     input[i] *= dq[i];
 
-  vp9_short_inv_walsh4x4_x8_c(input, output, 4 << 1);
+  vp9_short_iwalsh4x4_c(input, output, 4 << 1);
   vpx_memset(input, 0, 32);
   vp9_add_residual_4x4(output, pred, pitch, dest, stride);
 }
diff --git a/vp9/encoder/vp9_dct.c b/vp9/encoder/vp9_dct.c
index bb6e44fea..6365ed9a2 100644
--- a/vp9/encoder/vp9_dct.c
+++ b/vp9/encoder/vp9_dct.c
@@ -553,7 +553,7 @@ void vp9_short_fht8x8_c(int16_t *input, int16_t *output,
   }
 }
 
-void vp9_short_walsh4x4_x8_c(short *input, short *output, int pitch) {
+void vp9_short_walsh4x4_c(short *input, short *output, int pitch) {
   int i;
   int a1, b1, c1, d1;
   short *ip = input;
@@ -593,9 +593,9 @@ void vp9_short_walsh4x4_x8_c(short *input, short *output, int pitch) {
   }
 }
 
-void vp9_short_walsh8x4_x8_c(short *input, short *output, int pitch) {
-  vp9_short_walsh4x4_x8_c(input,   output,    pitch);
-  vp9_short_walsh4x4_x8_c(input + 4, output + 16, pitch);
+void vp9_short_walsh8x4_c(short *input, short *output, int pitch) {
+  vp9_short_walsh4x4_c(input,   output,    pitch);
+  vp9_short_walsh4x4_c(input + 4, output + 16, pitch);
 }
 
 
diff --git a/vp9/encoder/vp9_encodeframe.c b/vp9/encoder/vp9_encodeframe.c
index dd2004b01..428e585e1 100644
--- a/vp9/encoder/vp9_encodeframe.c
+++ b/vp9/encoder/vp9_encodeframe.c
@@ -1214,10 +1214,10 @@ static void init_encode_frame_mb_context(VP9_COMP *cpi) {
 
 static void switch_lossless_mode(VP9_COMP *cpi, int lossless) {
   if (lossless) {
-    cpi->mb.fwd_txm8x4            = vp9_short_walsh8x4_x8;
-    cpi->mb.fwd_txm4x4            = vp9_short_walsh4x4_x8;
-    cpi->mb.e_mbd.inv_txm4x4_1    = vp9_short_inv_walsh4x4_1_x8;
-    cpi->mb.e_mbd.inv_txm4x4      = vp9_short_inv_walsh4x4_x8;
+    cpi->mb.fwd_txm8x4            = vp9_short_walsh8x4;
+    cpi->mb.fwd_txm4x4            = vp9_short_walsh4x4;
+    cpi->mb.e_mbd.inv_txm4x4_1    = vp9_short_iwalsh4x4_1;
+    cpi->mb.e_mbd.inv_txm4x4      = vp9_short_iwalsh4x4;
     cpi->mb.optimize              = 0;
     cpi->common.filter_level      = 0;
     cpi->zbin_mode_boost_enabled  = FALSE;
@@ -1225,8 +1225,8 @@ static void switch_lossless_mode(VP9_COMP *cpi, int lossless) {
   } else {
     cpi->mb.fwd_txm8x4            = vp9_short_fdct8x4;
     cpi->mb.fwd_txm4x4            = vp9_short_fdct4x4;
-    cpi->mb.e_mbd.inv_txm4x4_1    = vp9_short_idct4x4llm_1;
-    cpi->mb.e_mbd.inv_txm4x4      = vp9_short_idct4x4llm;
+    cpi->mb.e_mbd.inv_txm4x4_1    = vp9_short_idct4x4_1;
+    cpi->mb.e_mbd.inv_txm4x4      = vp9_short_idct4x4;
   }
 }
 
diff --git a/vp9/encoder/vp9_onyx_if.c b/vp9/encoder/vp9_onyx_if.c
index c3353cee9..8674fc0c3 100644
--- a/vp9/encoder/vp9_onyx_if.c
+++ b/vp9/encoder/vp9_onyx_if.c
@@ -843,8 +843,8 @@ void vp9_set_speed_features(VP9_COMP *cpi) {
   cpi->mb.fwd_txm8x4    = vp9_short_fdct8x4;
   cpi->mb.fwd_txm4x4    = vp9_short_fdct4x4;
   if (cpi->oxcf.lossless || cpi->mb.e_mbd.lossless) {
-    cpi->mb.fwd_txm8x4    = vp9_short_walsh8x4_x8;
-    cpi->mb.fwd_txm4x4    = vp9_short_walsh4x4_x8;
+    cpi->mb.fwd_txm8x4    = vp9_short_walsh8x4;
+    cpi->mb.fwd_txm4x4    = vp9_short_walsh4x4;
   }
 
   cpi->mb.quantize_b_4x4      = vp9_regular_quantize_b_4x4;
@@ -1217,11 +1217,11 @@ void vp9_change_config(VP9_PTR ptr, VP9_CONFIG *oxcf) {
 
   cpi->oxcf.lossless = oxcf->lossless;
   if (cpi->oxcf.lossless) {
-    cpi->mb.e_mbd.inv_txm4x4_1    = vp9_short_inv_walsh4x4_1_x8;
-    cpi->mb.e_mbd.inv_txm4x4      = vp9_short_inv_walsh4x4_x8;
+    cpi->mb.e_mbd.inv_txm4x4_1    = vp9_short_iwalsh4x4_1;
+    cpi->mb.e_mbd.inv_txm4x4      = vp9_short_iwalsh4x4;
   } else {
-    cpi->mb.e_mbd.inv_txm4x4_1    = vp9_short_idct4x4llm_1;
-    cpi->mb.e_mbd.inv_txm4x4      = vp9_short_idct4x4llm;
+    cpi->mb.e_mbd.inv_txm4x4_1    = vp9_short_idct4x4_1;
+    cpi->mb.e_mbd.inv_txm4x4      = vp9_short_idct4x4;
   }
 
   cpi->baseline_gf_interval = DEFAULT_GF_INTERVAL;
diff --git a/vp9/vp9_common.mk b/vp9/vp9_common.mk
index f330b464a..9ee6f8599 100644
--- a/vp9/vp9_common.mk
+++ b/vp9/vp9_common.mk
@@ -28,7 +28,7 @@ VP9_COMMON_SRCS-yes += common/vp9_filter.c
 VP9_COMMON_SRCS-yes += common/vp9_filter.h
 VP9_COMMON_SRCS-yes += common/vp9_findnearmv.c
 VP9_COMMON_SRCS-yes += common/generic/vp9_systemdependent.c
-VP9_COMMON_SRCS-yes += common/vp9_idctllm.c
+VP9_COMMON_SRCS-yes += common/vp9_idct.c
 VP9_COMMON_SRCS-yes += common/vp9_alloccommon.h
 VP9_COMMON_SRCS-yes += common/vp9_blockd.h
 VP9_COMMON_SRCS-yes += common/vp9_common.h
@@ -91,7 +91,7 @@ VP9_COMMON_SRCS-$(CONFIG_POSTPROC) += common/vp9_postproc.c
 VP9_COMMON_SRCS-$(HAVE_MMX) += common/x86/vp9_iwalsh_mmx.asm
 VP9_COMMON_SRCS-$(HAVE_MMX) += common/x86/vp9_recon_mmx.asm
 VP9_COMMON_SRCS-$(HAVE_MMX) += common/x86/vp9_loopfilter_mmx.asm
-VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_idctllm_sse2.asm
+VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_idct_sse2.asm
 VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_iwalsh_sse2.asm
 VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_loopfilter_sse2.asm
 VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_recon_sse2.asm
@@ -110,13 +110,13 @@ VP9_COMMON_SRCS-yes += common/vp9_maskingmv.c
 VP9_COMMON_SRCS-$(HAVE_SSE3) += common/x86/vp9_mask_sse3.asm
 endif
 
-VP9_COMMON_SRCS-$(ARCH_X86)$(ARCH_X86_64) += common/x86/vp9_idctllm_x86.c
+VP9_COMMON_SRCS-$(ARCH_X86)$(ARCH_X86_64) += common/x86/vp9_idct_x86.c
 VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_sadmxn_x86.c
 ifeq ($(HAVE_SSE2),yes)
-vp9/common/x86/vp9_idctllm_x86.c.o: CFLAGS += -msse2
+vp9/common/x86/vp9_idct_x86.c.o: CFLAGS += -msse2
 vp9/common/x86/vp9_loopfilter_x86.c.o: CFLAGS += -msse2
 vp9/common/x86/vp9_sadmxn_x86.c.o: CFLAGS += -msse2
-vp9/common/x86/vp9_idctllm_x86.c.d: CFLAGS += -msse2
+vp9/common/x86/vp9_idct_x86.c.d: CFLAGS += -msse2
 vp9/common/x86/vp9_loopfilter_x86.c.d: CFLAGS += -msse2
 vp9/common/x86/vp9_sadmxn_x86.c.d: CFLAGS += -msse2
 endif