From patchwork Mon Feb 20 15:33:20 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: James Darnley X-Patchwork-Id: 2612 Delivered-To: ffmpegpatchwork@gmail.com Received: by 10.103.89.22 with SMTP id n22csp11970vsb; Mon, 20 Feb 2017 07:41:01 -0800 (PST) X-Received: by 10.28.144.66 with SMTP id s63mr20342512wmd.134.1487605261425; Mon, 20 Feb 2017 07:41:01 -0800 (PST) Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id b31si24334786wrd.314.2017.02.20.07.40.59; Mon, 20 Feb 2017 07:41:01 -0800 (PST) Received-SPF: pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) client-ip=79.124.17.100; Authentication-Results: mx.google.com; dkim=neutral (body hash did not verify) header.i=@ob-encoder-com.20150623.gappssmtp.com; spf=pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) smtp.mailfrom=ffmpeg-devel-bounces@ffmpeg.org Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id A32A2689C51; Mon, 20 Feb 2017 17:40:49 +0200 (EET) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-wm0-f65.google.com (mail-wm0-f65.google.com [74.125.82.65]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 616CA689B8E for ; Mon, 20 Feb 2017 17:40:43 +0200 (EET) Received: by mail-wm0-f65.google.com with SMTP id v77so15031669wmv.0 for ; Mon, 20 Feb 2017 07:40:51 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ob-encoder-com.20150623.gappssmtp.com; s=20150623; h=sender:from:to:subject:date:message-id:in-reply-to:references; bh=FYuLyQyCjVRkxzF2DUseY3ptgWFqHTGbrM1ror/erOs=; b=D9kJwYu4aEOmgJMsUygJhYpBTwLOp6Af7rEs1OtC+XtFhBqCIcbaoT0TC2OpqqX2PY j8t4yfApvUIPS3sbr9OR0nvgLGnoJUs3mixV7v4zqaG1kkG8zKk570hFpJ7TrgE4DmA+ ///O7MIoCMRVu0VLFn+buemAD4giFwOhmwA9GGqj98ZCk5FHTfMZ1wCEIrToMCPSVNIp BiAQrJZchB+Dc01ZjYIJ0ClrpmliX50T+/RSzEez6xvl5pE7wmKWfS+L57u33fe+2O8R qYiTvisZFB66ZItZTzsfcvzsOuIXHGcZWz4eQTbyGtd98LS5G/XLaznK/2Ag3m+BG7Bn jlIQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:sender:from:to:subject:date:message-id :in-reply-to:references; bh=FYuLyQyCjVRkxzF2DUseY3ptgWFqHTGbrM1ror/erOs=; b=UV1hCco2IF5N9KUp4iz2PEojpVRF5z/2BymmeiKdnknI5WsrtofGGJ42Rys+r6d2l8 /1HjQ4Po2lhIUhATzeJLMpI3y238dy8ax4x2sCOiQz7vwzJiKxvrS2Nd0zb+MHsK1N3O cCY6BgDs1APGDn+8WmCK/5xd48u+xfjzRitL6221mpEuEUyh7NcuPp7rZU6cAWlkHM5z CT0vAzgdaUhR9VUpkqpFPblicIv5DhUY9yy2izJIWm9A3ZGjL7q750JN9ha6WgjLcwAj Q4F5qRLTQzomNoGO7bZluT7RlHTyyAh2B5WWSPQMEH6Ph0tarWcIuIC8fs/0XW7v4VnZ rx1w== X-Gm-Message-State: AMke39k2CZf3NC4I+4qc9QTR/40wZmIHpoFASavU64kko3KJpiFzFeVmiaRUUgNBj6ZC+w== X-Received: by 10.28.111.155 with SMTP id c27mr19051853wmi.18.1487604833478; Mon, 20 Feb 2017 07:33:53 -0800 (PST) Received: from localhost.localdomain (d51A44418.access.telenet.be. [81.164.68.24]) by smtp.gmail.com with ESMTPSA id c133sm13865977wmd.13.2017.02.20.07.33.52 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 20 Feb 2017 07:33:52 -0800 (PST) From: James Darnley To: FFmpeg development discussions and patches Date: Mon, 20 Feb 2017 16:33:20 +0100 Message-Id: <20170220153324.1732-3-jdarnley@obe.tv> X-Mailer: git-send-email 2.8.3 In-Reply-To: <20170220153324.1732-1-jdarnley@obe.tv> References: <20170220153324.1732-1-jdarnley@obe.tv> Subject: [FFmpeg-devel] [PATCH 2/6] avcodec/h264: add avx 8-bit 4:2:0 chroma h deblock/loop filter X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: FFmpeg development discussions and patches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Reply-To: FFmpeg development discussions and patches MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" ~1.14x faster (93 vs. 81 cycles) compared with mmxext function --- libavcodec/x86/h264_deblock.asm | 70 +++++++++++++++++++++++++++++++++++++++++ libavcodec/x86/h264dsp_init.c | 3 ++ 2 files changed, 73 insertions(+) diff --git a/libavcodec/x86/h264_deblock.asm b/libavcodec/x86/h264_deblock.asm index 2e84ca3..0465c9f 100644 --- a/libavcodec/x86/h264_deblock.asm +++ b/libavcodec/x86/h264_deblock.asm @@ -1059,6 +1059,54 @@ ff_chroma_intra_body_mmxext: paddb m2, m6 ret +%macro LOAD_8_ROWS 8 + movd m0, %1 + movd m1, %2 + movd m2, %3 + movd m3, %4 + movd m4, %5 + movd m5, %6 + movd m6, %7 + movd m7, %8 +%endmacro + +%macro STORE_8_ROWS 8 + movd %1, m0 + movd %2, m1 + movd %3, m2 + movd %4, m3 + movd %5, m4 + movd %6, m5 + movd %7, m6 + movd %8, m7 +%endmacro + +%macro TRANSPOSE_8x4B_XMM 0 + punpcklbw m0, m1 + punpcklbw m2, m3 + punpcklbw m4, m5 + punpcklbw m6, m7 + punpcklwd m0, m2 + punpcklwd m4, m6 + punpckhdq m2, m0, m4 + punpckldq m0, m4 + MOVHL m1, m0 + MOVHL m3, m2 +%endmacro + +%macro TRANSPOSE_4x8B_XMM 0 + punpcklbw m0, m1 + punpcklbw m2, m3 + punpckhwd m4, m0, m2 + punpcklwd m0, m2 + MOVHL m6, m4 + MOVHL m2, m0 + pshufd m1, m0, 1 + pshufd m3, m2, 1 + pshufd m5, m4, 1 + pshufd m7, m6, 1 +%endmacro + %macro CHROMA_INTER_BODY_XMM 1 LOAD_MASK alpha_d, beta_d movd m6, [tc0_q] @@ -1078,6 +1126,15 @@ ff_chroma_intra_body_mmxext: sub %1, stride_q %endmacro +%macro CHROMA_H_START_XMM 2 + movsxdifnidn stride_q, stride_d + dec alpha_d + dec beta_d + lea %2, [3*stride_q] + mov %1, pix_q + add %1, %2 +%endmacro + %macro DEBLOCK_CHROMA_XMM 1 INIT_XMM %1 @@ -1093,6 +1150,19 @@ cglobal deblock_v_chroma_8, 5, 6, 8, pix_, stride_, alpha_, beta_, tc0_ movq [pix_q], m2 RET +cglobal deblock_h_chroma_8, 5, 7, 8, 0-16, pix_, stride_, alpha_, beta_, tc0_ + CHROMA_H_START_XMM r5, r6 + LOAD_8_ROWS PASS8ROWS(pix_q - 2, r5 - 2, stride_q, r6) + TRANSPOSE_8x4B_XMM + movq [rsp], m0 + movq [rsp + 8], m3 + CHROMA_INTER_BODY_XMM 1 + movq m0, [rsp] + movq m3, [rsp + 8] + TRANSPOSE_4x8B_XMM + STORE_8_ROWS PASS8ROWS(pix_q - 2, r5 - 2, stride_q, r6) +RET + %endmacro ; DEBLOCK_CHROMA_XMM DEBLOCK_CHROMA_XMM avx diff --git a/libavcodec/x86/h264dsp_init.c b/libavcodec/x86/h264dsp_init.c index 6794aa5..0b15471 100644 --- a/libavcodec/x86/h264dsp_init.c +++ b/libavcodec/x86/h264dsp_init.c @@ -319,6 +319,9 @@ av_cold void ff_h264dsp_init_x86(H264DSPContext *c, const int bit_depth, #endif c->h264_v_loop_filter_chroma = ff_deblock_v_chroma_8_avx; + if (chroma_format_idc <= 1) { + c->h264_h_loop_filter_chroma = ff_deblock_h_chroma_8_avx; + } } } else if (bit_depth == 10) { if (EXTERNAL_MMXEXT(cpu_flags)) {