[x265] [PATCH] asm: intrapred dc8 sse2 high bit
chen
chenm003 at 163.com
Fri Feb 27 05:50:15 CET 2015
在 2015-02-27 05:36:45，dave <dtyx265 at gmail.com> 写道：
On 02/25/2015 05:59 AM, chen wrote:
At 2015-02-25 08:16:44,dtyx265 at gmail.com wrote:
># HG changeset patch
># User David T Yuen <dtyx265 at gmail.com>
># Date 1424823199 28800
># Node ID bd1d713da87bb1e3022c80f462398bd78a95ce48
># Parent 644d27ca0b197455393171ba705b1190f3d9b420
>asm: intrapred dc8 sse2 high bit
>
>This replaces c code for systems using ssse3 to sse2 processors
>The code is backported from intrapred dc8 sse4 high bit
>
>./test/TestBench --testbench intrapred | grep 8x8
>intra_dc_8x8[f=0] 1.62x 443.94 717.32
>intra_dc_8x8[f=1] 1.59x 729.98 1157.66
>
>diff -r 644d27ca0b19 -r bd1d713da87b source/common/x86/asm-primitives.cpp
>--- a/source/common/x86/asm-primitives.cpp Tue Feb 24 14:47:01 2015 -0800
>+++ b/source/common/x86/asm-primitives.cpp Tue Feb 24 16:13:19 2015 -0800
>@@ -864,6 +864,7 @@
> ALL_LUMA_TU_S(transpose, transpose, sse2);
>
> p.cu[BLOCK_4x4].intra_pred[DC_IDX] = x265_intra_pred_dc4_sse2;
>+ p.cu[BLOCK_8x8].intra_pred[DC_IDX] = x265_intra_pred_dc8_sse2;
>
> p.cu[BLOCK_4x4].sse_ss = x265_pixel_ssd_ss_4x4_mmx2;
> ALL_LUMA_CU(sse_ss, pixel_ssd_ss, sse2);
>diff -r 644d27ca0b19 -r bd1d713da87b source/common/x86/intrapred16.asm
>--- a/source/common/x86/intrapred16.asm Tue Feb 24 14:47:01 2015 -0800
>+++ b/source/common/x86/intrapred16.asm Tue Feb 24 16:13:19 2015 -0800
>@@ -86,6 +86,7 @@
> cextern pw_8
> cextern pw_16
> cextern pw_1023
>+cextern pd_2
> cextern pd_16
> cextern pd_32
> cextern pw_4096
>@@ -166,6 +167,89 @@
> ;-----------------------------------------------------------------------------------
> ; void intra_pred_dc(pixel* dst, intptr_t dstStride, pixel* above, int, int filter)
> ;-----------------------------------------------------------------------------------
>+INIT_XMM sse2
>+cglobal intra_pred_dc8, 5, 8, 2
>+ add r2, 2
>+ add r1, r1
>+ movu m0, [r2 + 32]
>+ movu m1, [r2]
>+
>+ paddw m0, m1
>+ movhlps m1, m0
>+ paddw m0, m1
>+ pshufd m1, m0, 1
>+ paddw m0, m1
>+ pmaddwd m0, [pw_1]
>+
>+ paddw m0, [pw_8]
>+ psraw m0, 4 ; sum = sum / 16
>+ pshuflw m0, m0, 0
>+ pshufd m0, m0, 0 ; m0 = word [dc_val ...]
>+
>+ test r4d, r4d
>+
>+ ; store DC 8x8
>+ lea r6, [r1 + r1 * 2]
>+ lea r5, [r6 + r1 * 2]
>+ lea r7, [r6 + r1 * 4]
>+ movu [r0], m0
>+ movu [r0 + r1], m0
>+ movu [r0 + r1 * 2], m0
>+ movu [r0 + r6], m0
>+ movu [r0 + r1 * 4], m0
>+ movu [r0 + r5], m0
>+ movu [r0 + r6 * 2], m0
>+ movu [r0 + r7], m0
>+
>+ ; Do DC Filter
>+ jz .end
>+ mova m1, [pd_2]
>+ pmullw m1, m0
[x x x x] * [0 2 0 2] = [0 dc*2 0 dc*2], am I right?
Why not PSLL?
For some reason using psllw in place of pmullw hurts performance of unfiltered in 8 and 10 bit far more than you would expect for such a small change. 8 bit speed is cut by more than half
[MC] we need use IACA to analyze performance, in Agner's doc, PMULL latency are 5 cycles, and through is 1 cycles, PADD just 1 and 0.5
./test/TestBench --testbench intrapred | grep 8x8
intra_dc_8x8[f=0] 1.84x 492.58 906.63
intra_dc_8x8[f=1] 2.38x 537.68 1280.02
and 10 bit is barely faster than c code.
./test/TestBench --testbench intrapred | grep 8x8
intra_dc_8x8[f=0] 1.06x 677.54 719.80
intra_dc_8x8[f=1] 1.60x 725.05 1157.54
I will resubmit with other adjustments
>+ paddw m1, [pd_2]
>+ movd r4d, m1 ; r4d = DC * 2 + 2
MOVD fix previous issue
>+ paddw m1, m0 ; m1 = DC * 3 + 2
>+ pshuflw m1, m1, 0
>+ pshufd m1, m1, 0 ; m1 = pixDCx3
>+
>+ ; filter top
>+ movu m0, [r2]
>+ paddw m0, m1
>+ psraw m0, 2
>+ movu [r0], m0
>+
>+ ; filter top-left
>+ movzx r3d, word [r2 + 32]
>+ add r4d, r3d
>+ movzx r3d, word [r2]
>+ add r3d, r4d
>+ shr r3d, 2
>+ mov [r0], r3w
>+
>+ ; filter left
>+ movu m0, [r2 + 34]
>+ paddw m0, m1
>+ psraw m0, 2
>+ movh r3, m0
>+ mov [r0 + r1], r3w
>+ shr r3, 16
>+ mov [r0 + r1 * 2], r3w
>+ shr r3, 16
>+ mov [r0 + r6], r3w
>+ shr r3, 16
>+ mov [r0 + r1 * 4], r3w
>+ pshufd m0, m0, 0x6E
>+ movh r3, m0
>+ mov [r0 + r5], r3w
>+ shr r3, 16
>+ mov [r0 + r6 * 2], r3w
>+ shr r3, 16
>+ mov [r0 + r7], r3w
>+.end:
>+ RET
>+
>+;-----------------------------------------------------------------------------------
>+; void intra_pred_dc(pixel* dst, intptr_t dstStride, pixel* above, int, int filter)
>+;-----------------------------------------------------------------------------------
> INIT_XMM sse4
> cglobal intra_pred_dc4, 5,6,2
> lea r3, [r2 + 18]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20150227/0d9a0c39/attachment-0001.html>
More information about the x265-devel
mailing list