From patchwork Fri Mar 11 07:52:09 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Wu, Jianhua" X-Patchwork-Id: 34661 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6838:3486:0:0:0:0 with SMTP id ek6csp1471760nkb; Thu, 10 Mar 2022 23:52:37 -0800 (PST) X-Google-Smtp-Source: ABdhPJwS9kUnbGKhc+8ItMG5OombeeRXXeUNkVdPVxEN09cdgTF82arH+gfmSjE310zEBnysyIEF X-Received: by 2002:a17:907:50a1:b0:6da:9f90:a55f with SMTP id fv33-20020a17090750a100b006da9f90a55fmr7676298ejc.36.1646985157403; Thu, 10 Mar 2022 23:52:37 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1646985157; cv=none; d=google.com; s=arc-20160816; b=NSJ5ELRpgPHqI2Z7koDGz1wtGgq/Lxf1CgcP6S1Hr9p25nl3nU3Ms2lgR7MCvGBpul o8wVhGI7NCv0hjI14xu8W4T2U1Q57SsB4MAjzzJuMe0cQCjqlC/KpKz/CPsOGj4bNc0i DycHL/EP1kZ1G6NUBK7klvd8TYn/nxhrBsSxVoVG9eaqtvGxlwkf6hLmL2YuV50B7i+E YiIpA+BdLk/QnZTtjcjiZtWhdeRNwcNLsh2SIEU/iH8ZxkuRYuI5T7yPdJZeWpMuvKwm yynfycW94uSa04Z+4/7nlv9mPBtBJW4P58ndvrCx9hPV7TdD1oTUmSh813XXauBCwSKH AYfA== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:mime-version:cc:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:message-id:date:to:from:dkim-signature :delivered-to; bh=nBTEoc4UDRbD3LzEDxxE6q7Sta6pkWBvtS8LB/rV7Uk=; b=uULs+WrJz1bkR+pnqiPrs9DTMIqyEf31AuKJ581DEgmGZiElqj2b8LmVG6rwo3JNW0 mLYQrBr4uw/aZB6FW1/vCu7tRLp1UinxeEMQm7ZS2mt+O4dxeHhJheuXmXR/m+NPxa5o JnLthBhqQ1LP1VIfT/crUqsHVwSMBcmU0Ka9VNiH+rupoznpJkn7plU17cwUHlg6J7xE 3vUA+fMTVPaN0AaZ0X6m3nVqYgkGXZS1DI3zFq57WDQPSaU45gqYt7q0lb3tInBwseVA 8LuIBeOQZuLKPkY+lsvvsN0jcAdJR5zBRE6rwVf5TYk2ej75GNeyUo3jrqVfQOzcqrGi BeeA== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@intel.com header.s=Intel header.b=k7t4sFuG; 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 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id bf4-20020a0564021a4400b0041602994aaesi4226837edb.33.2022.03.10.23.52.36; Thu, 10 Mar 2022 23:52:37 -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=@intel.com header.s=Intel header.b=k7t4sFuG; 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 80B7468B155; Fri, 11 Mar 2022 09:52:31 +0200 (EET) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id A6B2E68AF69 for ; Fri, 11 Mar 2022 09:52:24 +0200 (EET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1646985149; x=1678521149; h=from:to:cc:subject:date:message-id; bh=ccTQy+SIdhOwJjAjeL0Dz6qCKRwpl9uNJPKanYYGI6o=; b=k7t4sFuGwLruSt2XxwMf9Tq/fk/5+hOOcGy823Q+2Kjf5NKs5rbsV7kC XbYXBOgMQL787NygjF5uE8RD63Q12XllYh22b9HpNrj7hLYlXKJyBDq0G DrMpdSywrRC9SSQ1zGmFGO8nbTHqebvqHq/+Cnf3bCILqDXQfq7Sz9Ps+ OGtmugK+vD4u9J+1p9KqjfTZ/WO5qa6Nf56ySc/+PID8HzBoM2TkSaFWL vmFJNuHRmJOcC2QqMcs6VpuIfZMDTv8DBA1lMHPl3d/3XMvjnfa/0hiBr vqbaFBCS6OpoUk6lEaasUYbJaUBqiqb33cdkhIlbw4UuZsGJmhp53ioQZ w==; X-IronPort-AV: E=McAfee;i="6200,9189,10282"; a="341950691" X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="341950691" Received: from orsmga005.jf.intel.com ([10.7.209.41]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 10 Mar 2022 23:52:20 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="712767070" Received: from otc-skl-e5-server.sh.intel.com ([10.239.43.106]) by orsmga005.jf.intel.com with ESMTP; 10 Mar 2022 23:52:19 -0800 From: jianhua.wu-at-intel.com@ffmpeg.org To: ffmpeg-devel@ffmpeg.org Date: Fri, 11 Mar 2022 15:52:09 +0800 Message-Id: <20220311075213.49925-1-jianhua.wu@intel.com> X-Mailer: git-send-email 2.17.1 Subject: [FFmpeg-devel] [PATCH v2 1/5] avcodec/x86/hevc_mc: add qpel_h8_8_avx512icl and qpel_hv8_8_avx512icl X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.29 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 Cc: Wu Jianhua MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: m2hVTh9H6b6C From: Wu Jianhua This commit uses the instruction `vpdpbusd` introduced by AVX512 VNNI to calculate the horizontal filter. ff_hevc_put_hevc_qpel_h8_8_sse4 1039169 ff_hevc_put_hevc_qpel_h8_8_avx512icl 677153 ff_hevc_put_hevc_qpel_hv8_8_sse4 3603511 ff_hevc_put_hevc_qpel_hv8_8_avx512icl 2995354 Signed-off-by: Wu Jianhua --- libavcodec/x86/hevc_mc.asm | 137 ++++++++++++++++++++++++++++++++++ libavcodec/x86/hevcdsp.h | 3 + libavcodec/x86/hevcdsp_init.c | 4 + 3 files changed, 144 insertions(+) diff --git a/libavcodec/x86/hevc_mc.asm b/libavcodec/x86/hevc_mc.asm index ff6ed0711a..52fa3ec948 100644 --- a/libavcodec/x86/hevc_mc.asm +++ b/libavcodec/x86/hevc_mc.asm @@ -87,6 +87,26 @@ QPEL_TABLE 12, 4, w, sse4 QPEL_TABLE 8,16, b, avx2 QPEL_TABLE 10, 8, w, avx2 +QPEL_TABLE 8, 1, b, avx512icl_h +QPEL_TABLE 8, 1, d, avx512icl_v + +pb_qpel_shuffle_index: db 0, 1, 2, 3 + db 1, 2, 3, 4 + db 2, 3, 4, 5 + db 3, 4, 5, 6 + db 4, 5, 6, 7 + db 5, 6, 7, 8 + db 6, 7, 8, 9 + db 7, 8, 9, 10 + db 4, 5, 6, 7 + db 5, 6, 7, 8 + db 6, 7, 8, 9 + db 7, 8, 9, 10 + db 8, 9, 10, 11 + db 9, 10, 11, 12 + db 10, 11, 12, 13 + db 11, 12, 13, 14 + SECTION .text %define MAX_PB_SIZE 64 @@ -1670,3 +1690,120 @@ HEVC_PUT_HEVC_QPEL_HV 16, 10 %endif ;AVX2 %endif ; ARCH_X86_64 + +%macro QPEL_FILTER_H 5 +%define %%table hevc_qpel_filters_avx512icl_h_%1 +%assign %%offset 4 + dec %2q + shl %2q, 3 +%ifdef PIC + lea %5q, [%%table] + %define FILTER %5q +%else + %define FILTER %%table +%endif + vpbroadcastd m%3, [FILTER + %2q + 0*%%offset] + vpbroadcastd m%4, [FILTER + %2q + 1*%%offset] +%endmacro + +%macro QPEL_FILTER_V 5 + vpbroadcastd m%3, [%5 + %2q + 4*%4] +%endmacro + +%macro QPEL_LOAD_SHUF 2 + movu m%1, [pb_qpel_shuffle_index + 0] + movu m%2, [pb_qpel_shuffle_index + 32] +%endmacro + +; required: m0-m5 +; %1: dst register index +; %2: name for src +%macro QPEL_H_LOAD_COMPUTE 2 + pxor m%1, m%1 + movu xm4, [%2q - 3] + vpermb m5, m2, m4 + vpermb m4, m3, m4 + vpdpbusd m%1, m5, m0 + vpdpbusd m%1, m4, m1 +%endmacro + +%macro HEVC_PUT_HEVC_QPEL_AVX512ICL 2 +cglobal hevc_put_hevc_qpel_h%1_%2, 5, 6, 8, dst, src, srcstride, height, mx, tmp + QPEL_FILTER_H %1, mx, 0, 1, tmp + QPEL_LOAD_SHUF 2, 3 +.loop: + QPEL_H_LOAD_COMPUTE 6, src + vpmovdw [dstq], m6 + LOOP_END dst, src, srcstride + RET +%endmacro + +%macro HEVC_PUT_HEVC_QPEL_HV_AVX512ICL 2 +cglobal hevc_put_hevc_qpel_hv%1_%2, 6, 7, 27, dst, src, srcstride, height, mx, my, tmp +%assign %%shift 6 +%assign %%extra 7 + QPEL_FILTER_H %1, mx, 0, 1, tmp + QPEL_LOAD_SHUF 2, 3 + lea tmpq, [srcstrideq*3] + sub srcq, tmpq + sub myq, 1 + shl myq, 5 +%ifdef PIC +%define %%table hevc_qpel_filters_avx512icl_v_%1 + lea tmpq, [%%table] + %define FILTER tmpq +%else + %define FILTER %%table +%endif +%assign %%i 6 +%assign %%j 0 +%rep %1 + QPEL_FILTER_V %1, my, %%i, %%j, FILTER + %assign %%i %%i+1 + %assign %%j %%j+1 +%endrep +%rep %%extra + QPEL_H_LOAD_COMPUTE %%i, src + add srcq, srcstrideq +%assign %%i %%i+1 +%endrep +.loop: + QPEL_H_LOAD_COMPUTE %%i, src + vpmulld m22, m14, m6 + vpmulld m23, m15, m7 + vpmulld m24, m16, m8 + vpmulld m25, m17, m9 + vpaddd m26, m22, m23 + vpaddd m24, m25 + vpaddd m26, m24 + vpmulld m22, m18, m10 + vpmulld m23, m19, m11 + vpmulld m24, m20, m12 + vpmulld m25, m21, m13 + vpaddd m22, m22, m23 + vpaddd m24, m25 + vpaddd m26, m24 + vpaddd m22, m26 + mova m14, m15 + mova m15, m16 + mova m16, m17 + mova m17, m18 + mova m18, m19 + mova m19, m20 + mova m20, m21 + vpsrad m22, %%shift + vpmovdw [dstq], m22 + LOOP_END dst, src, srcstride + + RET +%endmacro + +%if ARCH_X86_64 +%if HAVE_AVX512ICL_EXTERNAL + +INIT_YMM avx512icl +HEVC_PUT_HEVC_QPEL_AVX512ICL 8, 8 +HEVC_PUT_HEVC_QPEL_HV_AVX512ICL 8, 8 + +%endif +%endif diff --git a/libavcodec/x86/hevcdsp.h b/libavcodec/x86/hevcdsp.h index 67be0a9059..5a495d2563 100644 --- a/libavcodec/x86/hevcdsp.h +++ b/libavcodec/x86/hevcdsp.h @@ -233,6 +233,9 @@ WEIGHTING_PROTOTYPES(8, sse4); WEIGHTING_PROTOTYPES(10, sse4); WEIGHTING_PROTOTYPES(12, sse4); +void ff_hevc_put_hevc_qpel_h8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); +void ff_hevc_put_hevc_qpel_hv8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); + /////////////////////////////////////////////////////////////////////////////// // TRANSFORM_ADD /////////////////////////////////////////////////////////////////////////////// diff --git a/libavcodec/x86/hevcdsp_init.c b/libavcodec/x86/hevcdsp_init.c index 8a3fa2744b..e51375ebc2 100644 --- a/libavcodec/x86/hevcdsp_init.c +++ b/libavcodec/x86/hevcdsp_init.c @@ -878,6 +878,10 @@ void ff_hevc_dsp_init_x86(HEVCDSPContext *c, const int bit_depth) c->add_residual[3] = ff_hevc_add_residual_32_8_avx2; } + if (EXTERNAL_AVX512ICL(cpu_flags) && ARCH_X86_64) { + c->put_hevc_qpel[3][0][1] = ff_hevc_put_hevc_qpel_h8_8_avx512icl; + c->put_hevc_qpel[3][1][1] = ff_hevc_put_hevc_qpel_hv8_8_avx512icl; + } } else if (bit_depth == 10) { if (EXTERNAL_MMXEXT(cpu_flags)) { c->add_residual[0] = ff_hevc_add_residual_4_10_mmxext; From patchwork Fri Mar 11 07:52:10 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Wu, Jianhua" X-Patchwork-Id: 34662 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6838:3486:0:0:0:0 with SMTP id ek6csp1471867nkb; Thu, 10 Mar 2022 23:52:49 -0800 (PST) X-Google-Smtp-Source: ABdhPJxLcNgALrF2E+INyI477OnvQKFvIWHG5x++ggPRw5CAbsgbOctXJ9vVuBMOlfF3hpeep29y X-Received: by 2002:a17:907:94ca:b0:6da:b785:f067 with SMTP id dn10-20020a17090794ca00b006dab785f067mr7284148ejc.654.1646985169146; Thu, 10 Mar 2022 23:52:49 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1646985169; cv=none; d=google.com; s=arc-20160816; b=XLwcJd8624xEAuLeRLsFbpsCrvOHYc7fO/zEd7jzX12iU4dKX33o+h/GwUX+t+tpZn WAVz45CYVUi/p+jsms6PDX3S8BZ43ATfoFExYfdv+5ulROX7sJF/b2+Q/ZpYqo6wWZ9D 0BljATs4iumuEXhzXDAcenKbg4MsUrYYqdz2Brf0yXdaotQHuNENziclEatr3vV4SR6M 7Rg8iQEzQww7S6Gd+PIQMMEaFK7DZLX1KtqxiALfZrLgzVFCkWTDOlPphG7h9WDBxTIQ KUmekSQN+SkK+uxwsn5i62rqh+gx5OgbK9L6/mgX0c8EEqf/+tG6xwBEsRayDSF2+zsv aewg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:mime-version:cc:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:references:in-reply-to:message-id:date :to:from:dkim-signature:delivered-to; bh=W++d2Br0wM1tbg3AZJAuPCpJyudU0RjzKKI4GOB6YU8=; b=W5Dg9LdtyXgX1COIv0PNo8OkS8c+GSel83uG9XavzU39i5tiYvcbAFf+zYaR/+J2nX HCSDv1I/MlcO5whiUhQj81bCyMrZfxQmLccBRKgqpEZtO66XpivjA2YlXW0dNZMY6ynL 0/F3VUFnLJ5z4JTKYhH+y0mgNaYSb+3QZj3PbA+fd/JiftDlOYpszKVtosECmMqLn8Qk obO8I1Zlk9T4gyyGqjVhfm69OG+2o4nLKHY3r3LIYC7487LCacv+MG3dRfKeqWAgU4IG 5PAuzYg3aXUV5QBv132Y/nikmahknjC8SSS+AhLpaABf7phaXLwTFoqpwusxD702troY DRow== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@intel.com header.s=Intel header.b=BRYt2TMu; 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 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id ne16-20020a1709077b9000b006da8847bca1si5458138ejc.201.2022.03.10.23.52.48; Thu, 10 Mar 2022 23:52:49 -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=@intel.com header.s=Intel header.b=BRYt2TMu; 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 9CAD068B165; Fri, 11 Mar 2022 09:52:34 +0200 (EET) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 3ADB268B165 for ; Fri, 11 Mar 2022 09:52:27 +0200 (EET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1646985152; x=1678521152; h=from:to:cc:subject:date:message-id:in-reply-to: references; bh=6MAIRR7xGuN+AFGjPgIeZAk1pIPiy48gLt4rfZ57duA=; b=BRYt2TMuEmUrALP+1Dzptjt1h6K+kHFqtFTKZcbYhNyNnHtQa6h1lDhz XtcCZLA8n3KmmUrKp+Zsk0RYh3SrNj0IDbrP4gYmUaqkeV1pp1hz6Xsgn QwPJC+/MC6449dxcwYOFoECPf4qQTl1IhDaEQyvt+rr9R7v2212HQSGKI sfQIchz9yPVqS6eJcuq2h3qbrMO+DREx8v00vBwok3EOwzUQDL/MdpLiM cs5Xmi3YqTrlYkpa2y/6S3zZcqxhngObh2yuhV4md9WTJDsYnpRotgNl7 yqF+2VXK2oSHoo2RTATkcYAN2Qxz9Vu4mVSf1TfY2u5UDErsiZauJC7dj g==; X-IronPort-AV: E=McAfee;i="6200,9189,10282"; a="341950693" X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="341950693" Received: from orsmga005.jf.intel.com ([10.7.209.41]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 10 Mar 2022 23:52:21 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="712767077" Received: from otc-skl-e5-server.sh.intel.com ([10.239.43.106]) by orsmga005.jf.intel.com with ESMTP; 10 Mar 2022 23:52:20 -0800 From: jianhua.wu-at-intel.com@ffmpeg.org To: ffmpeg-devel@ffmpeg.org Date: Fri, 11 Mar 2022 15:52:10 +0800 Message-Id: <20220311075213.49925-2-jianhua.wu@intel.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20220311075213.49925-1-jianhua.wu@intel.com> References: <20220311075213.49925-1-jianhua.wu@intel.com> Subject: [FFmpeg-devel] [PATCH v2 2/5] avcodec/x86/hevc_mc: add qpel_h16_8_avx512icl X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.29 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 Cc: Wu Jianhua MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: Mq6DpFsqcRuo From: Wu Jianhua ff_hevc_put_hevc_qpel_h16_8_sse4 3290870 ff_hevc_put_hevc_qpel_h16_8_avx512icl 1730033 Signed-off-by: Wu Jianhua --- libavcodec/x86/hevc_mc.asm | 26 +++++++++++++++++++++++++- libavcodec/x86/hevcdsp.h | 1 + libavcodec/x86/hevcdsp_init.c | 1 + 3 files changed, 27 insertions(+), 1 deletion(-) diff --git a/libavcodec/x86/hevc_mc.asm b/libavcodec/x86/hevc_mc.asm index 52fa3ec948..4e39cdd7fe 100644 --- a/libavcodec/x86/hevc_mc.asm +++ b/libavcodec/x86/hevc_mc.asm @@ -89,6 +89,7 @@ QPEL_TABLE 10, 8, w, avx2 QPEL_TABLE 8, 1, b, avx512icl_h QPEL_TABLE 8, 1, d, avx512icl_v +QPEL_TABLE 16, 1, b, avx512icl_h pb_qpel_shuffle_index: db 0, 1, 2, 3 db 1, 2, 3, 4 @@ -98,6 +99,14 @@ pb_qpel_shuffle_index: db 0, 1, 2, 3 db 5, 6, 7, 8 db 6, 7, 8, 9 db 7, 8, 9, 10 + db 8, 9, 10, 11 + db 9, 10, 11, 12 + db 10, 11, 12, 13 + db 11, 12, 13, 14 + db 12, 13, 14, 15 + db 13, 14, 15, 16 + db 14, 15, 16, 17 + db 15, 16, 17, 18 db 4, 5, 6, 7 db 5, 6, 7, 8 db 6, 7, 8, 9 @@ -106,6 +115,14 @@ pb_qpel_shuffle_index: db 0, 1, 2, 3 db 9, 10, 11, 12 db 10, 11, 12, 13 db 11, 12, 13, 14 + db 12, 13, 14, 15 + db 13, 14, 15, 16 + db 14, 15, 16, 17 + db 15, 16, 17, 18 + db 16, 17, 18, 19 + db 17, 18, 19, 20 + db 18, 19, 20, 21 + db 19, 20, 21, 22 SECTION .text @@ -1712,7 +1729,7 @@ HEVC_PUT_HEVC_QPEL_HV 16, 10 %macro QPEL_LOAD_SHUF 2 movu m%1, [pb_qpel_shuffle_index + 0] - movu m%2, [pb_qpel_shuffle_index + 32] + movu m%2, [pb_qpel_shuffle_index + 64] %endmacro ; required: m0-m5 @@ -1720,7 +1737,11 @@ HEVC_PUT_HEVC_QPEL_HV 16, 10 ; %2: name for src %macro QPEL_H_LOAD_COMPUTE 2 pxor m%1, m%1 +%if mmsize == 64 + movu ym4, [%2q - 3] +%else movu xm4, [%2q - 3] +%endif vpermb m5, m2, m4 vpermb m4, m3, m4 vpdpbusd m%1, m5, m0 @@ -1805,5 +1826,8 @@ INIT_YMM avx512icl HEVC_PUT_HEVC_QPEL_AVX512ICL 8, 8 HEVC_PUT_HEVC_QPEL_HV_AVX512ICL 8, 8 +INIT_ZMM avx512icl +HEVC_PUT_HEVC_QPEL_AVX512ICL 16, 8 + %endif %endif diff --git a/libavcodec/x86/hevcdsp.h b/libavcodec/x86/hevcdsp.h index 5a495d2563..6e3fc01ad0 100644 --- a/libavcodec/x86/hevcdsp.h +++ b/libavcodec/x86/hevcdsp.h @@ -234,6 +234,7 @@ WEIGHTING_PROTOTYPES(10, sse4); WEIGHTING_PROTOTYPES(12, sse4); void ff_hevc_put_hevc_qpel_h8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); +void ff_hevc_put_hevc_qpel_h16_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_hv8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); /////////////////////////////////////////////////////////////////////////////// diff --git a/libavcodec/x86/hevcdsp_init.c b/libavcodec/x86/hevcdsp_init.c index e51375ebc2..a73c2e2bd9 100644 --- a/libavcodec/x86/hevcdsp_init.c +++ b/libavcodec/x86/hevcdsp_init.c @@ -880,6 +880,7 @@ void ff_hevc_dsp_init_x86(HEVCDSPContext *c, const int bit_depth) } if (EXTERNAL_AVX512ICL(cpu_flags) && ARCH_X86_64) { c->put_hevc_qpel[3][0][1] = ff_hevc_put_hevc_qpel_h8_8_avx512icl; + c->put_hevc_qpel[5][0][1] = ff_hevc_put_hevc_qpel_h16_8_avx512icl; c->put_hevc_qpel[3][1][1] = ff_hevc_put_hevc_qpel_hv8_8_avx512icl; } } else if (bit_depth == 10) { From patchwork Fri Mar 11 07:52:11 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Wu, Jianhua" X-Patchwork-Id: 34663 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6838:3486:0:0:0:0 with SMTP id ek6csp1471932nkb; Thu, 10 Mar 2022 23:53:00 -0800 (PST) X-Google-Smtp-Source: ABdhPJwfilB8CkfMhqwXnpBgoPfj9b3+n+NPEzde31zbpOZx/ipU4rt3kwsuz4yxHSac+WLy1MAE X-Received: by 2002:a17:906:b1d0:b0:6cf:7f39:50e0 with SMTP id bv16-20020a170906b1d000b006cf7f3950e0mr7700295ejb.760.1646985180119; Thu, 10 Mar 2022 23:53:00 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1646985180; cv=none; d=google.com; s=arc-20160816; b=cr5xnX44pRejfLP1LJQNkHEPseazkEJWiqH+NhTFz5B0KgaPAFiEZ6TFK00p9uwjFi I6yEf6IxTCLEDi6ID422i180w4T9CtxazkwIvFMIU6qJ8wCIvuxTNMLtAtlZNjXHQOd1 f4Yh3MrMvOc7bsU3zWZg7D0dGy7XqvGVMZhpp2PwuwbauXMqk6X2GKwRdnb0BfrIazfB UhnB4IpS6SVZh5SU7TrdiH64cuIuSpOx67bcwXvCkg6T6OFPcdRl111B/MbB/liptuVe ZJNfPUnGgp+l8/tTKdiQZndO0LTsu0GqzP8IrlNtY8es5910rcFJ4kJxpHVG6pGMz5aP w6Cg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:mime-version:cc:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:references:in-reply-to:message-id:date :to:from:dkim-signature:delivered-to; bh=GwjY9cjjC8MlWFA1brRARLsWcxYJWrdIdzWQ/mghacw=; b=qDgLF91LJ/dtg4iSvc1Q7jycsypyAwv6ZY3NireOPJPAAPxLFZPZEmQMm9bHCpCZdo ptGGTvG3xMK0zpvbmJoaDvJ/IWHyGjq3nUu4Z0bQAH6Ttjt1L8T8wexfUy2UlDGW6ePD yOR3EOBl55kS300xGH/qGAKGIG7q9NbuMi2OcJK23E+ulW/cTJeR7CegnUPJEzZXFHRs Tg8S4kdNUdEcxKKg1Bb4A0viLuTfbfijIPC9axrFJsumJKzOPpLwBlLFiYLQhku8pCGX P7oa45ZrKdIUFICSycsTVUwbckl/qPLrtbcxa2pQgJxj8M2C77QfCmapuYuv89kzyA10 RV+w== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@intel.com header.s=Intel header.b=I0NVOOMa; 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 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id v29-20020a50a45d000000b0041593c729a5si4483486edb.468.2022.03.10.23.52.59; Thu, 10 Mar 2022 23:53:00 -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=@intel.com header.s=Intel header.b=I0NVOOMa; 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 9183168B1A5; Fri, 11 Mar 2022 09:52:37 +0200 (EET) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id A9FF568B174 for ; Fri, 11 Mar 2022 09:52:30 +0200 (EET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1646985155; x=1678521155; h=from:to:cc:subject:date:message-id:in-reply-to: references; bh=UJ/jvzL4qYRDYw3MwyLVaKyJnGSTwxA+B1NZ8jsxto0=; b=I0NVOOMak0XsxNRRkH92qNpRchfDbuTLX9+ERaZypEzS74UZ0FVTpI8i 3qFliaKeEs3WU6H5IFXAdjnYmPkIC77yvntJWxvSZFXUES0zjmd/dKEkw 2BCSwPMABn+8gQac6rCeC9P0R81TedWunwaY64mhat6HELMOx1XhEjdsM +tR3GzA6B2cMMzrfzQtcs94ZT1lux9GRB34sbgmu31hfAZfNPrk12kljJ z3/idccKYJ068Oln48rrOIHusAZ8hIp+7MmdMWcurigdKeWxa1GkoKNyu EYKgZhgUxiHLtDAG67RZx1ueoGEAbX+ADIFj6FTj8fT4F6z8XBE/D5cA0 g==; X-IronPort-AV: E=McAfee;i="6200,9189,10282"; a="341950695" X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="341950695" Received: from orsmga005.jf.intel.com ([10.7.209.41]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 10 Mar 2022 23:52:22 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="712767082" Received: from otc-skl-e5-server.sh.intel.com ([10.239.43.106]) by orsmga005.jf.intel.com with ESMTP; 10 Mar 2022 23:52:21 -0800 From: jianhua.wu-at-intel.com@ffmpeg.org To: ffmpeg-devel@ffmpeg.org Date: Fri, 11 Mar 2022 15:52:11 +0800 Message-Id: <20220311075213.49925-3-jianhua.wu@intel.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20220311075213.49925-1-jianhua.wu@intel.com> References: <20220311075213.49925-1-jianhua.wu@intel.com> Subject: [FFmpeg-devel] [PATCH v2 3/5] avcodec/x86/hevc_mc: add qpel_h4_8_avx512icl X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.29 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 Cc: Wu Jianhua MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: 4VzJCk5zKhHT From: Wu Jianhua ff_hevc_put_hevc_qpel_h4_8_sse4 993694 ff_hevc_put_hevc_qpel_h4_8_avx512icl 686647 Signed-off-by: Wu Jianhua --- libavcodec/x86/hevc_mc.asm | 9 +++++++++ libavcodec/x86/hevcdsp.h | 1 + libavcodec/x86/hevcdsp_init.c | 1 + 3 files changed, 11 insertions(+) diff --git a/libavcodec/x86/hevc_mc.asm b/libavcodec/x86/hevc_mc.asm index 4e39cdd7fe..ff59ae3509 100644 --- a/libavcodec/x86/hevc_mc.asm +++ b/libavcodec/x86/hevc_mc.asm @@ -87,6 +87,7 @@ QPEL_TABLE 12, 4, w, sse4 QPEL_TABLE 8,16, b, avx2 QPEL_TABLE 10, 8, w, avx2 +QPEL_TABLE 4, 1, b, avx512icl_h QPEL_TABLE 8, 1, b, avx512icl_h QPEL_TABLE 8, 1, d, avx512icl_v QPEL_TABLE 16, 1, b, avx512icl_h @@ -1754,7 +1755,12 @@ cglobal hevc_put_hevc_qpel_h%1_%2, 5, 6, 8, dst, src, srcstride, height, mx, tmp QPEL_LOAD_SHUF 2, 3 .loop: QPEL_H_LOAD_COMPUTE 6, src +%if %1 == 4 + vpmovdw xm6, m6 + movq [dstq], xm6 +%else vpmovdw [dstq], m6 +%endif LOOP_END dst, src, srcstride RET %endmacro @@ -1822,6 +1828,9 @@ cglobal hevc_put_hevc_qpel_hv%1_%2, 6, 7, 27, dst, src, srcstride, height, mx, m %if ARCH_X86_64 %if HAVE_AVX512ICL_EXTERNAL +INIT_XMM avx512icl +HEVC_PUT_HEVC_QPEL_AVX512ICL 4, 8 + INIT_YMM avx512icl HEVC_PUT_HEVC_QPEL_AVX512ICL 8, 8 HEVC_PUT_HEVC_QPEL_HV_AVX512ICL 8, 8 diff --git a/libavcodec/x86/hevcdsp.h b/libavcodec/x86/hevcdsp.h index 6e3fc01ad0..51ffdc9628 100644 --- a/libavcodec/x86/hevcdsp.h +++ b/libavcodec/x86/hevcdsp.h @@ -233,6 +233,7 @@ WEIGHTING_PROTOTYPES(8, sse4); WEIGHTING_PROTOTYPES(10, sse4); WEIGHTING_PROTOTYPES(12, sse4); +void ff_hevc_put_hevc_qpel_h4_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_h8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_h16_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_hv8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); diff --git a/libavcodec/x86/hevcdsp_init.c b/libavcodec/x86/hevcdsp_init.c index a73c2e2bd9..58b91459ed 100644 --- a/libavcodec/x86/hevcdsp_init.c +++ b/libavcodec/x86/hevcdsp_init.c @@ -879,6 +879,7 @@ void ff_hevc_dsp_init_x86(HEVCDSPContext *c, const int bit_depth) c->add_residual[3] = ff_hevc_add_residual_32_8_avx2; } if (EXTERNAL_AVX512ICL(cpu_flags) && ARCH_X86_64) { + c->put_hevc_qpel[1][0][1] = ff_hevc_put_hevc_qpel_h4_8_avx512icl; c->put_hevc_qpel[3][0][1] = ff_hevc_put_hevc_qpel_h8_8_avx512icl; c->put_hevc_qpel[5][0][1] = ff_hevc_put_hevc_qpel_h16_8_avx512icl; c->put_hevc_qpel[3][1][1] = ff_hevc_put_hevc_qpel_hv8_8_avx512icl; From patchwork Fri Mar 11 07:52:12 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Wu, Jianhua" X-Patchwork-Id: 34664 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6838:3486:0:0:0:0 with SMTP id ek6csp1472028nkb; Thu, 10 Mar 2022 23:53:11 -0800 (PST) X-Google-Smtp-Source: ABdhPJz6LWffRV63eQ66KZxqgn0UAKhIQwXCSGTW9DMXGMRiFgHLdQixe+jFth+i0Zk388EeCUtT X-Received: by 2002:a50:e1ca:0:b0:413:b403:f8e6 with SMTP id m10-20020a50e1ca000000b00413b403f8e6mr7637217edl.204.1646985190871; Thu, 10 Mar 2022 23:53:10 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1646985190; cv=none; d=google.com; s=arc-20160816; b=tHKkIF/fFBtOoPe+ULSIdqDtIfSFoYdq1qjf+YxKlD6dWxsodJVKQGhjl56AouCjFE ywfYH0/22NJqYtG/3B2pBb1UH3vZrtxMgsduHVX3kZxxhYjQPZQdDkgx34t/bSHLQG2k AQBqdaO7qokReWVche0UJeZC/dnm7ZTz5zf+9ryAwimA0kDy2/vB1hWkMc0qXNYFUczc UeEPt3ceuHRIazIFSL1OuOsCJy82zlw2Cm6hODh0jvMTTRkB7vH/W492YLHm/3+KG7nx A1ANALT/UAVC1l5JVEWWhZ2HifXp7lKD5ZtAe2BQf1eFMUGS11b2dDg0iUxC0c5YmBxI gVNQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:mime-version:cc:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:references:in-reply-to:message-id:date :to:from:dkim-signature:delivered-to; bh=aPxLRMPREJPv1d/SiSCED3BQU+4rkksAtak8v5w2Rrk=; b=xB0a8qXVDnBzctq7usTHo84cL+HrCe1mw9vZDgUTQDP0TrSYf+/3W/uhqW+l1qXKMF pFii/gtiDUzI9SGpKko5+o+jkF1C+4VJWYvxc01k0tp4c9be3U8g7uNajh8Oq3X36PiU C0wQkW1h3X1eA85y5CpDB6JB/xk2TACiNpGqb7xo2htO9oyu/yCTmU0sJDiSz9+NPMYg 8xVIdHEzc7xzk2SUMqDs1FZK+2PayT13RFpknvabIA2zRtkRwbkK6bDoPnLEqrTc6Oeb ZFDxkronPCNucfLjKTrRg7JMmtWBP5HOwQyvBcNFxYirRHMj9V6He0w8ip4K6zgnPxug EGCg== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@intel.com header.s=Intel header.b=jsG7V5AL; 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 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id hh14-20020a170906a94e00b006d0ce01ed85si4343949ejb.163.2022.03.10.23.53.10; Thu, 10 Mar 2022 23:53:10 -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=@intel.com header.s=Intel header.b=jsG7V5AL; 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 C449F68B1C8; Fri, 11 Mar 2022 09:52:40 +0200 (EET) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 0024668B176 for ; Fri, 11 Mar 2022 09:52:31 +0200 (EET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1646985158; x=1678521158; h=from:to:cc:subject:date:message-id:in-reply-to: references; bh=ogWCuNAdUA4k38UY8KLszuft1oUrH1ccJ439XKAXxp0=; b=jsG7V5ALE0uOgG2ZYLo6+/0cxWEaJOIQL7011Alhxu2lS+1QrhT6JSM3 D5jDhsDYHXysMa3xmulCVT2frZuC5RURU78ojCEZtr9VJAUiATmOKavPN /JGXiN4qZFBBwPkqiBB5hXUqg6khXbCd86pAGaXscOMXLQ6Ogun/Nw1/2 4PYKSGVoP7/00bk8RqrctwLMHq2DxLW8xHlty5O1QZaf5485r+VQE2MyE oer6+PDwXMk/1Zuv02gFSkXQQWusJCsv8rvR1yyIsckytTWJ8WfD3m/eY 9ylX3R9bkWZrwgGkHeJ18trcWS9AiB+5IiwloSLNiGMl2n455INr2xFdG g==; X-IronPort-AV: E=McAfee;i="6200,9189,10282"; a="341950696" X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="341950696" Received: from orsmga005.jf.intel.com ([10.7.209.41]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 10 Mar 2022 23:52:24 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="712767087" Received: from otc-skl-e5-server.sh.intel.com ([10.239.43.106]) by orsmga005.jf.intel.com with ESMTP; 10 Mar 2022 23:52:22 -0800 From: jianhua.wu-at-intel.com@ffmpeg.org To: ffmpeg-devel@ffmpeg.org Date: Fri, 11 Mar 2022 15:52:12 +0800 Message-Id: <20220311075213.49925-4-jianhua.wu@intel.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20220311075213.49925-1-jianhua.wu@intel.com> References: <20220311075213.49925-1-jianhua.wu@intel.com> Subject: [FFmpeg-devel] [PATCH v2 4/5] avcodec/x86/hevc_mc: add qpel_h32_8_avx512icl X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.29 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 Cc: Wu Jianhua MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: SBrcFa/8Aoiu From: Wu Jianhua ff_hevc_put_hevc_qpel_h32_8_sse4 14122151 ff_hevc_put_hevc_qpel_h32_8_avx2 9337675 ff_hevc_put_hevc_qpel_h32_8_avx512icl 6424654 Signed-off-by: Wu Jianhua --- libavcodec/x86/hevc_mc.asm | 17 ++++++++++++++--- libavcodec/x86/hevcdsp.h | 1 + libavcodec/x86/hevcdsp_init.c | 1 + 3 files changed, 16 insertions(+), 3 deletions(-) diff --git a/libavcodec/x86/hevc_mc.asm b/libavcodec/x86/hevc_mc.asm index ff59ae3509..fc3faac376 100644 --- a/libavcodec/x86/hevc_mc.asm +++ b/libavcodec/x86/hevc_mc.asm @@ -91,6 +91,7 @@ QPEL_TABLE 4, 1, b, avx512icl_h QPEL_TABLE 8, 1, b, avx512icl_h QPEL_TABLE 8, 1, d, avx512icl_v QPEL_TABLE 16, 1, b, avx512icl_h +QPEL_TABLE 32, 1, b, avx512icl_h pb_qpel_shuffle_index: db 0, 1, 2, 3 db 1, 2, 3, 4 @@ -1736,12 +1737,17 @@ HEVC_PUT_HEVC_QPEL_HV 16, 10 ; required: m0-m5 ; %1: dst register index ; %2: name for src -%macro QPEL_H_LOAD_COMPUTE 2 +; %3: optional offset +%macro QPEL_H_LOAD_COMPUTE 2-3 +%assign %%offset 0 +%if %0 == 3 +%assign %%offset %3 +%endif pxor m%1, m%1 %if mmsize == 64 - movu ym4, [%2q - 3] + movu ym4, [%2q + %%offset - 3] %else - movu xm4, [%2q - 3] + movu xm4, [%2q + %%offset - 3] %endif vpermb m5, m2, m4 vpermb m4, m3, m4 @@ -1760,6 +1766,10 @@ cglobal hevc_put_hevc_qpel_h%1_%2, 5, 6, 8, dst, src, srcstride, height, mx, tmp movq [dstq], xm6 %else vpmovdw [dstq], m6 +%endif +%if %1 == 32 + QPEL_H_LOAD_COMPUTE 7, src, 16 + vpmovdw [dstq + 32], m7 %endif LOOP_END dst, src, srcstride RET @@ -1837,6 +1847,7 @@ HEVC_PUT_HEVC_QPEL_HV_AVX512ICL 8, 8 INIT_ZMM avx512icl HEVC_PUT_HEVC_QPEL_AVX512ICL 16, 8 +HEVC_PUT_HEVC_QPEL_AVX512ICL 32, 8 %endif %endif diff --git a/libavcodec/x86/hevcdsp.h b/libavcodec/x86/hevcdsp.h index 51ffdc9628..8d3c3cc75f 100644 --- a/libavcodec/x86/hevcdsp.h +++ b/libavcodec/x86/hevcdsp.h @@ -236,6 +236,7 @@ WEIGHTING_PROTOTYPES(12, sse4); void ff_hevc_put_hevc_qpel_h4_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_h8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_h16_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); +void ff_hevc_put_hevc_qpel_h32_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_hv8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); /////////////////////////////////////////////////////////////////////////////// diff --git a/libavcodec/x86/hevcdsp_init.c b/libavcodec/x86/hevcdsp_init.c index 58b91459ed..ce5902c693 100644 --- a/libavcodec/x86/hevcdsp_init.c +++ b/libavcodec/x86/hevcdsp_init.c @@ -882,6 +882,7 @@ void ff_hevc_dsp_init_x86(HEVCDSPContext *c, const int bit_depth) c->put_hevc_qpel[1][0][1] = ff_hevc_put_hevc_qpel_h4_8_avx512icl; c->put_hevc_qpel[3][0][1] = ff_hevc_put_hevc_qpel_h8_8_avx512icl; c->put_hevc_qpel[5][0][1] = ff_hevc_put_hevc_qpel_h16_8_avx512icl; + c->put_hevc_qpel[7][0][1] = ff_hevc_put_hevc_qpel_h32_8_avx512icl; c->put_hevc_qpel[3][1][1] = ff_hevc_put_hevc_qpel_hv8_8_avx512icl; } } else if (bit_depth == 10) { From patchwork Fri Mar 11 07:52:13 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Wu, Jianhua" X-Patchwork-Id: 34665 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6838:3486:0:0:0:0 with SMTP id ek6csp1472094nkb; Thu, 10 Mar 2022 23:53:20 -0800 (PST) X-Google-Smtp-Source: ABdhPJzZE0IhV6gep4SH5Z+oTjigyf727Jstdl5mYVt0p4o5vrSgB5XDsNU+tWnYD3TfzeJJH0EP X-Received: by 2002:a17:906:a210:b0:6d5:9fa:11ce with SMTP id r16-20020a170906a21000b006d509fa11cemr7979408ejy.172.1646985200784; Thu, 10 Mar 2022 23:53:20 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1646985200; cv=none; d=google.com; s=arc-20160816; b=UvRZr/E0MP2SAam7w9PLk9RHg/+q3BRDTND/fnSDQ+J8mXfFoFguLlw0f9HUmDLbWt Uf+7hJV0Z92Z/LtZE7Og6s6INJ6nicHe9adI4KSeT27ymQUIHIxsJT5qss8V2iMdlzQV ATWj8V69FP4OtOKI6xeB3lVBszeIqc6xW6zx0jdYYF/T16QQI2lm2GGyoVvZrojs5NEh +5ZkUaRSb/8w0KDiGvcAcTLzUcYNgQDO0UAfNRGZFpspHvQClgQZjq9FTh4SpFwJGTlZ ODZ92co9qW15yyaElbkGZolDZx463S6YFGTPXEN4fpTSK50O5NEdjvECm5LNpOJuGqz9 6CiA== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:mime-version:cc:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:references:in-reply-to:message-id:date :to:from:dkim-signature:delivered-to; bh=m/bDLVLpDddS+2U7NaSxh5Evo/w6oz8urE/8AkNPp7Y=; b=jlrSt4P+N3j1unFG9zPpoYY8FKfQ3I2ge/EeCgWaphj3tNUJJ2Z0cVV19rHQPPeovF 6Qt2ei31QteVVB+76llbdOiEoyiOT+GL/WAXSLhDLS2oT+phG2sOtRct1uR+8/DLAHhr JdZoxdpDp+niuDa2O/zq4Psv/uA0JTDJPux8+2DHR27GoyH0bPmmHlNTGzsbrA0V1+8F /VmcbmIeEI+qRhu1Ukkhmb2xyV6RTLPj9rtQQtXvPO1/55WhKki/XCaz2TjuMCVvnwsr +aKu3d1xCzXfLUu3g3XOvg9aELbHM+p25baURlzvZOdyvf0EBJ6F3zBtJiql9qqxvygC GD3w== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@intel.com header.s=Intel header.b=fiOGuhBA; 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 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id q27-20020a1709066b1b00b00697f95a67ffsi4437689ejr.578.2022.03.10.23.53.20; Thu, 10 Mar 2022 23:53:20 -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=@intel.com header.s=Intel header.b=fiOGuhBA; 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 BC63868B1D4; Fri, 11 Mar 2022 09:52:41 +0200 (EET) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 2EFE368B1AD for ; Fri, 11 Mar 2022 09:52:32 +0200 (EET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1646985158; x=1678521158; h=from:to:cc:subject:date:message-id:in-reply-to: references; bh=p8Up5C9s98CuM4nJVHMW0wrL/0muKvLFXbscob50wLs=; b=fiOGuhBAwWjKdygoFrF9DOYSglHV1N4oksYn2WNxDDomTMT7TFmn62Y3 xzqt70uWCxsDCf08sRcaqRSv0sgjZh5jUmn4FZYfADWb2UGU7YGBXR08x uvsVAEm3K1RUcrvCntKxI1jmAPyy9vZPzExQxlYer9ytb8ZEYyh4dazJL Ov1Mbyur0OqWYDE919UKqbEg9R2z4QkuGiUDqT9RHqD4RUp6En41eXpiP 0DYwn9J5ArrcI+MvmaruEYSa9V0yRZYkpmGeBlDOhGx8/zn2DxXxu2CkY QehsaXAcia6ozPCDVZ1jTjION0SUXElVAOUBOcwcDkKpNlKwKPUcZOIJW Q==; X-IronPort-AV: E=McAfee;i="6200,9189,10282"; a="341950697" X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="341950697" Received: from orsmga005.jf.intel.com ([10.7.209.41]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 10 Mar 2022 23:52:25 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.90,173,1643702400"; d="scan'208";a="712767094" Received: from otc-skl-e5-server.sh.intel.com ([10.239.43.106]) by orsmga005.jf.intel.com with ESMTP; 10 Mar 2022 23:52:24 -0800 From: jianhua.wu-at-intel.com@ffmpeg.org To: ffmpeg-devel@ffmpeg.org Date: Fri, 11 Mar 2022 15:52:13 +0800 Message-Id: <20220311075213.49925-5-jianhua.wu@intel.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20220311075213.49925-1-jianhua.wu@intel.com> References: <20220311075213.49925-1-jianhua.wu@intel.com> Subject: [FFmpeg-devel] [PATCH v2 5/5] avcodec/x86/hevc_mc: add qpel_h64_8_avx512icl X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.29 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 Cc: Wu Jianhua MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: PWMooPgpLwly From: Wu Jianhua ff_hevc_put_hevc_qpel_h64_8_sse4 56782981 ff_hevc_put_hevc_qpel_h64_8_avx2 40097816 ff_hevc_put_hevc_qpel_h64_8_avx512icl 25488576 Signed-off-by: Wu Jianhua --- libavcodec/x86/hevc_mc.asm | 10 +++++++++- libavcodec/x86/hevcdsp.h | 1 + libavcodec/x86/hevcdsp_init.c | 1 + 3 files changed, 11 insertions(+), 1 deletion(-) diff --git a/libavcodec/x86/hevc_mc.asm b/libavcodec/x86/hevc_mc.asm index fc3faac376..c07948dd9b 100644 --- a/libavcodec/x86/hevc_mc.asm +++ b/libavcodec/x86/hevc_mc.asm @@ -92,6 +92,7 @@ QPEL_TABLE 8, 1, b, avx512icl_h QPEL_TABLE 8, 1, d, avx512icl_v QPEL_TABLE 16, 1, b, avx512icl_h QPEL_TABLE 32, 1, b, avx512icl_h +QPEL_TABLE 64, 1, b, avx512icl_h pb_qpel_shuffle_index: db 0, 1, 2, 3 db 1, 2, 3, 4 @@ -1767,9 +1768,15 @@ cglobal hevc_put_hevc_qpel_h%1_%2, 5, 6, 8, dst, src, srcstride, height, mx, tmp %else vpmovdw [dstq], m6 %endif -%if %1 == 32 +%if %1 > 16 QPEL_H_LOAD_COMPUTE 7, src, 16 vpmovdw [dstq + 32], m7 +%endif +%if %1 > 32 + QPEL_H_LOAD_COMPUTE 6, src, 32 + QPEL_H_LOAD_COMPUTE 7, src, 48 + vpmovdw [dstq + 64], m6 + vpmovdw [dstq + 96], m7 %endif LOOP_END dst, src, srcstride RET @@ -1848,6 +1855,7 @@ HEVC_PUT_HEVC_QPEL_HV_AVX512ICL 8, 8 INIT_ZMM avx512icl HEVC_PUT_HEVC_QPEL_AVX512ICL 16, 8 HEVC_PUT_HEVC_QPEL_AVX512ICL 32, 8 +HEVC_PUT_HEVC_QPEL_AVX512ICL 64, 8 %endif %endif diff --git a/libavcodec/x86/hevcdsp.h b/libavcodec/x86/hevcdsp.h index 8d3c3cc75f..24e35bc032 100644 --- a/libavcodec/x86/hevcdsp.h +++ b/libavcodec/x86/hevcdsp.h @@ -237,6 +237,7 @@ void ff_hevc_put_hevc_qpel_h4_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t void ff_hevc_put_hevc_qpel_h8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_h16_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_h32_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); +void ff_hevc_put_hevc_qpel_h64_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); void ff_hevc_put_hevc_qpel_hv8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width); /////////////////////////////////////////////////////////////////////////////// diff --git a/libavcodec/x86/hevcdsp_init.c b/libavcodec/x86/hevcdsp_init.c index ce5902c693..f3061bda84 100644 --- a/libavcodec/x86/hevcdsp_init.c +++ b/libavcodec/x86/hevcdsp_init.c @@ -883,6 +883,7 @@ void ff_hevc_dsp_init_x86(HEVCDSPContext *c, const int bit_depth) c->put_hevc_qpel[3][0][1] = ff_hevc_put_hevc_qpel_h8_8_avx512icl; c->put_hevc_qpel[5][0][1] = ff_hevc_put_hevc_qpel_h16_8_avx512icl; c->put_hevc_qpel[7][0][1] = ff_hevc_put_hevc_qpel_h32_8_avx512icl; + c->put_hevc_qpel[9][0][1] = ff_hevc_put_hevc_qpel_h64_8_avx512icl; c->put_hevc_qpel[3][1][1] = ff_hevc_put_hevc_qpel_hv8_8_avx512icl; } } else if (bit_depth == 10) {