<div dir="ltr"><div class="gmail_default" style="font-family:trebuchet ms,sans-serif;font-size:small">sent new patch with modified commit message. thanks.<br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, May 20, 2015 at 12:13 PM, Deepthi Nandakumar <span dir="ltr"><<a href="mailto:deepthi@multicorewareinc.com" target="_blank">deepthi@multicorewareinc.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Thanks, there are already 2 patches with similar commit messages. Can you add more details to the commit message?<br></div><div class="gmail_extra"><br><div class="gmail_quote"><div><div class="h5">On Wed, May 20, 2015 at 12:04 PM, <span dir="ltr"><<a href="mailto:dnyaneshwar@multicorewareinc.com" target="_blank">dnyaneshwar@multicorewareinc.com</a>></span> wrote:<br></div></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div class="h5"># HG changeset patch<br>
# User Dnyaneshwar G <<a href="mailto:dnyaneshwar@multicorewareinc.com" target="_blank">dnyaneshwar@multicorewareinc.com</a>><br>
# Date 1432102991 -19800<br>
# Wed May 20 11:53:11 2015 +0530<br>
# Node ID 5244b9a0d9a20262c99801a42e346e0b3e07b315<br>
# Parent cdf14fea15a846f2deca436a8e057711607f41bf<br>
asm: removed duplicate constants<br>
<br>
diff -r cdf14fea15a8 -r 5244b9a0d9a2 source/common/x86/intrapred8.asm<br>
--- a/source/common/x86/intrapred8.asm Wed May 20 11:02:10 2015 +0530<br>
+++ b/source/common/x86/intrapred8.asm Wed May 20 11:53:11 2015 +0530<br>
@@ -664,14 +664,6 @@<br>
ALIGN 32<br>
;; (blkSize - 1 - x)<br>
pw_planar4_0: dw 3, 2, 1, 0, 3, 2, 1, 0<br>
-pw_planar4_1: dw 3, 3, 3, 3, 3, 3, 3, 3<br>
-pw_planar8_0: dw 7, 6, 5, 4, 3, 2, 1, 0<br>
-pw_planar8_1: dw 7, 7, 7, 7, 7, 7, 7, 7<br>
-pw_planar16_0: dw 15, 14, 13, 12, 11, 10, 9, 8<br>
-pw_planar16_1: dw 15, 15, 15, 15, 15, 15, 15, 15<br>
-pw_planar32_1: dw 31, 31, 31, 31, 31, 31, 31, 31<br>
-pw_planar32_L: dw 31, 30, 29, 28, 27, 26, 25, 24<br>
-pw_planar32_H: dw 23, 22, 21, 20, 19, 18, 17, 16<br>
<br>
ALIGN 32<br>
c_ang8_mode_13: db 9, 23, 9, 23, 9, 23, 9, 23, 9, 23, 9, 23, 9, 23, 9, 23, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14<br>
@@ -714,7 +706,9 @@<br>
SECTION .text<br>
<br>
cextern pw_2<br>
+cextern pw_3<br>
cextern pw_4<br>
+cextern pw_7<br>
cextern pw_8<br>
cextern pw_16<br>
cextern pw_15<br>
@@ -1151,7 +1145,7 @@<br>
pshufd m4, m4, 0xAA<br>
<br>
pmullw m3, [multi_2Row] ; (x + 1) * topRight<br>
- pmullw m0, m1, [pw_planar4_1] ; (blkSize - 1 - y) * above[x]<br>
+ pmullw m0, m1, [pw_3] ; (blkSize - 1 - y) * above[x]<br>
paddw m3, [pw_4]<br>
paddw m3, m4<br>
paddw m3, m0<br>
@@ -1212,7 +1206,7 @@<br>
pshufd m4, m4, 0x44<br>
<br>
pmullw m3, [multiL] ; (x + 1) * topRight<br>
- pmullw m0, m1, [pw_planar8_1] ; (blkSize - 1 - y) * above[x]<br>
+ pmullw m0, m1, [pw_7] ; (blkSize - 1 - y) * above[x]<br>
paddw m3, [pw_8]<br>
paddw m3, m4<br>
paddw m3, m0<br>
@@ -1226,7 +1220,7 @@<br>
pshufhw m5, m2, 0x55 * (%1 - 4)<br>
pshufd m5, m5, 0xAA<br>
%endif<br>
- pmullw m5, [pw_planar8_0]<br>
+ pmullw m5, [pw_planar16_mul + mmsize]<br>
paddw m5, m3<br>
psraw m5, 4<br>
packuswb m5, m5<br>
@@ -1269,8 +1263,8 @@<br>
<br>
pmullw m4, m3, [multiH] ; (x + 1) * topRight<br>
pmullw m3, [multiL] ; (x + 1) * topRight<br>
- pmullw m1, m2, [pw_planar16_1] ; (blkSize - 1 - y) * above[x]<br>
- pmullw m5, m7, [pw_planar16_1] ; (blkSize - 1 - y) * above[x]<br>
+ pmullw m1, m2, [pw_15] ; (blkSize - 1 - y) * above[x]<br>
+ pmullw m5, m7, [pw_15] ; (blkSize - 1 - y) * above[x]<br>
paddw m4, [pw_16]<br>
paddw m3, [pw_16]<br>
paddw m4, m6<br>
@@ -1308,8 +1302,8 @@<br>
paddw m4, m1<br>
lea r0, [r0 + r1]<br>
%endif<br>
- pmullw m0, m5, [pw_planar8_0]<br>
- pmullw m5, [pw_planar16_0]<br>
+ pmullw m0, m5, [pw_planar16_mul + mmsize]<br>
+ pmullw m5, [pw_planar16_mul]<br>
paddw m0, m4<br>
paddw m5, m3<br>
psraw m5, 5<br>
@@ -1369,7 +1363,7 @@<br>
mova m9, m11<br>
mova m10, m11<br>
<br>
- mova m12, [pw_planar32_1]<br>
+ mova m12, [pw_31]<br>
movh m4, [r2 + 1]<br>
punpcklbw m4, m7<br>
psubw m8, m4<br>
@@ -1394,10 +1388,10 @@<br>
pmullw m4, m12<br>
paddw m3, m4<br>
<br>
- mova m12, [pw_planar32_L]<br>
- mova m13, [pw_planar32_H]<br>
- mova m14, [pw_planar16_0]<br>
- mova m15, [pw_planar8_0]<br>
+ mova m12, [pw_planar32_mul]<br>
+ mova m13, [pw_planar32_mul + mmsize]<br>
+ mova m14, [pw_planar16_mul]<br>
+ mova m15, [pw_planar16_mul + mmsize]<br>
%macro PROCESS 1<br>
pmullw m5, %1, m12<br>
pmullw m6, %1, m13<br>
@@ -1480,33 +1474,33 @@<br>
punpcklbw m4, m7<br>
psubw m5, m6, m4<br>
mova [rsp + 0 * mmsize], m5<br>
- pmullw m4, [pw_planar32_1]<br>
+ pmullw m4, [pw_31]<br>
paddw m0, m4<br>
<br>
movh m4, [r2 + 9]<br>
punpcklbw m4, m7<br>
psubw m5, m6, m4<br>
mova [rsp + 1 * mmsize], m5<br>
- pmullw m4, [pw_planar32_1]<br>
+ pmullw m4, [pw_31]<br>
paddw m1, m4<br>
<br>
movh m4, [r2 + 17]<br>
punpcklbw m4, m7<br>
psubw m5, m6, m4<br>
mova [rsp + 2 * mmsize], m5<br>
- pmullw m4, [pw_planar32_1]<br>
+ pmullw m4, [pw_31]<br>
paddw m2, m4<br>
<br>
movh m4, [r2 + 25]<br>
punpcklbw m4, m7<br>
psubw m5, m6, m4<br>
mova [rsp + 3 * mmsize], m5<br>
- pmullw m4, [pw_planar32_1]<br>
+ pmullw m4, [pw_31]<br>
paddw m3, m4<br>
<br>
%macro PROCESS 1<br>
- pmullw m5, %1, [pw_planar32_L]<br>
- pmullw m6, %1, [pw_planar32_H]<br>
+ pmullw m5, %1, [pw_planar32_mul]<br>
+ pmullw m6, %1, [pw_planar32_mul + mmsize]<br>
paddw m5, m0<br>
paddw m6, m1<br>
psraw m5, 6<br>
@@ -1514,8 +1508,8 @@<br>
packuswb m5, m6<br>
movu [r0], m5<br>
<br>
- pmullw m5, %1, [pw_planar16_0]<br>
- pmullw %1, [pw_planar8_0]<br>
+ pmullw m5, %1, [pw_planar16_mul]<br>
+ pmullw %1, [pw_planar16_mul + mmsize]<br>
paddw m5, m2<br>
paddw %1, m3<br>
psraw m5, 6<br>
@@ -2476,7 +2470,7 @@<br>
pshufd m4, m4, 0xAA<br>
<br>
pmullw m3, [multi_2Row] ; (x + 1) * topRight<br>
- pmullw m0, m1, [pw_planar4_1] ; (blkSize - 1 - y) * above[x]<br>
+ pmullw m0, m1, [pw_3] ; (blkSize - 1 - y) * above[x]<br>
mova m6, [pw_planar4_0]<br>
paddw m3, [pw_4]<br>
paddw m3, m4<br>
@@ -2535,8 +2529,8 @@<br>
punpcklbw m4, m0 ; v_bottomLeft<br>
<br>
pmullw m3, [multiL] ; (x + 1) * topRight<br>
- pmullw m0, m1, [pw_planar8_1] ; (blkSize - 1 - y) * above[x]<br>
- mova m6, [pw_planar8_0]<br>
+ pmullw m0, m1, [pw_7] ; (blkSize - 1 - y) * above[x]<br>
+ mova m6, [pw_planar16_mul + mmsize]<br>
paddw m3, [pw_8]<br>
paddw m3, m4<br>
paddw m3, m0<br>
@@ -2588,8 +2582,8 @@<br>
<br>
pmullw m4, m3, [multiH] ; (x + 1) * topRight<br>
pmullw m3, [multiL] ; (x + 1) * topRight<br>
- pmullw m1, m2, [pw_planar16_1] ; (blkSize - 1 - y) * above[x]<br>
- pmullw m5, m7, [pw_planar16_1] ; (blkSize - 1 - y) * above[x]<br>
+ pmullw m1, m2, [pw_15] ; (blkSize - 1 - y) * above[x]<br>
+ pmullw m5, m7, [pw_15] ; (blkSize - 1 - y) * above[x]<br>
paddw m4, [pw_16]<br>
paddw m3, [pw_16]<br>
paddw m4, m6<br>
@@ -2620,8 +2614,8 @@<br>
%endif<br>
%endif<br>
%endif<br>
- pmullw m0, m5, [pw_planar8_0]<br>
- pmullw m5, [pw_planar16_0]<br>
+ pmullw m0, m5, [pw_planar16_mul + mmsize]<br>
+ pmullw m5, [pw_planar16_mul]<br>
paddw m0, m4<br>
paddw m5, m3<br>
paddw m3, m6<br>
@@ -2740,25 +2734,25 @@<br>
paddw m3, [pw_32]<br>
<br>
pmovzxbw m4, [r2 + 1]<br>
- pmullw m5, m4, [pw_planar32_1]<br>
+ pmullw m5, m4, [pw_31]<br>
paddw m0, m5<br>
psubw m5, m6, m4<br>
mova m8, m5<br>
<br>
pmovzxbw m4, [r2 + 9]<br>
- pmullw m5, m4, [pw_planar32_1]<br>
+ pmullw m5, m4, [pw_31]<br>
paddw m1, m5<br>
psubw m5, m6, m4<br>
mova m9, m5<br>
<br>
pmovzxbw m4, [r2 + 17]<br>
- pmullw m5, m4, [pw_planar32_1]<br>
+ pmullw m5, m4, [pw_31]<br>
paddw m2, m5<br>
psubw m5, m6, m4<br>
mova m10, m5<br>
<br>
pmovzxbw m4, [r2 + 25]<br>
- pmullw m5, m4, [pw_planar32_1]<br>
+ pmullw m5, m4, [pw_31]<br>
paddw m3, m5<br>
psubw m5, m6, m4<br>
mova m11, m5<br>
@@ -2769,8 +2763,8 @@<br>
pshufb m4, m7<br>
punpcklbw m4, m7<br>
<br>
- pmullw m5, m4, [pw_planar32_L]<br>
- pmullw m6, m4, [pw_planar32_H]<br>
+ pmullw m5, m4, [pw_planar32_mul]<br>
+ pmullw m6, m4, [pw_planar32_mul + mmsize]<br>
paddw m5, m0<br>
paddw m6, m1<br>
paddw m0, m8<br>
@@ -2780,8 +2774,8 @@<br>
packuswb m5, m6<br>
movu [r0], m5<br>
<br>
- pmullw m5, m4, [pw_planar16_0]<br>
- pmullw m4, [pw_planar8_0]<br>
+ pmullw m5, m4, [pw_planar16_mul]<br>
+ pmullw m4, [pw_planar16_mul + mmsize]<br>
paddw m5, m2<br>
paddw m4, m3<br>
paddw m2, m10<br></div></div>
_______________________________________________<br>
x265-devel mailing list<br>
<a href="mailto:x265-devel@videolan.org" target="_blank">x265-devel@videolan.org</a><br>
<a href="https://mailman.videolan.org/listinfo/x265-devel" target="_blank">https://mailman.videolan.org/listinfo/x265-devel</a><br>
</blockquote></div><br></div>
<br>_______________________________________________<br>
x265-devel mailing list<br>
<a href="mailto:x265-devel@videolan.org">x265-devel@videolan.org</a><br>
<a href="https://mailman.videolan.org/listinfo/x265-devel" target="_blank">https://mailman.videolan.org/listinfo/x265-devel</a><br>
<br></blockquote></div><br></div>