From patchwork Sat May 4 14:48:38 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: uk7b@foxmail.com X-Patchwork-Id: 48492 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:e68f:b0:1af:836d:81b3 with SMTP id mz15csp424758pzb; Sat, 4 May 2024 07:50:26 -0700 (PDT) X-Forwarded-Encrypted: i=2; AJvYcCUhFW1OH6/HyEHHOzsHjQfR+rqiO79Ouu0O6najEV7/gZN1RNORyu/gr6pThYUhUYgiYr1RyJDldlxERkxJeZlsfdht1fKRp7bZEw== X-Google-Smtp-Source: AGHT+IE4kB5GDTED/UGr7XfyUQgDmITS4EqPUTaytETteTNhUUeNwIsmgontdHALgiOLJF3hynQM X-Received: by 2002:a50:c047:0:b0:572:7b08:d497 with SMTP id u7-20020a50c047000000b005727b08d497mr3288921edd.17.1714834226586; Sat, 04 May 2024 07:50:26 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1714834226; cv=none; d=google.com; s=arc-20160816; b=qisNRK/0RdcwTe0W5y50GIVXBg7SQRQCHXU6PKHsf7BXa96EcNXAUP4YSc4EO0Dro2 l+jy8vlRIcQiYUVkKYIr6+OsmYKq9qx5Dk8MpWDzgKRJe0h0kVBwJlDFwghx/utSlJAY n73N9YoTy2zZWLx2IGtLiEfREwE37H8YWhoAwEX63+BvDmrwFlDGAJQs77yU5HkvKJ7o E1WPsmZv6foL1e6BQxOrxuRf39oNh3PsBYSqx16mAclR9GECkkQH5qA3/a7mjcdOdDmp wPUn7owzkhzVBQBXPPbdtaeBxDEZ+nRNfvEmt1kbsMy6Ke14pzGgjLD381RJiPeGzaik pUwQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:cc:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:mime-version:references:in-reply-to:date :to:from:message-id:dkim-signature:delivered-to; bh=zSNDeGI4UEwQIQ+YpwdkGvE6DV5x4ej7sqO15/DdLj0=; fh=D0bFwGkf4X22/D/bfeDVrXKIx7S6kcXsNzy10j8ORbQ=; b=y7FekaMe8ukuIKeGUo4wuS1N5iGKewl1Xdd1oJ7AiskSIhLuYNHThb49cpS7GeWxnQ Tv0onNzeMDKnXyaSVoV+Ixx7IYMQHISgtgPtgo+bd5QJPu1waj1fv9otH91dwukDqJ40 PwZzlsM71kNNu4jPNxkTWcFEk9hFOE35R+ZJfgVDDwalpi0wN3AvLlCSkevyyUtKbDCQ 72TkgkdjHeXAXFJ6M78vIKZMGnEr3zR69G3fiPIx+cjIaIj0CuR/LKHx5kR+K4ZMopuh SVnU8DGmRHtXm0MqDXTrE+0urUfT4Y/cvebTW09mBN0pqh3rgffrJYvHg4LAmQmqLftO 6Q3A==; dara=google.com ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@foxmail.com header.s=s201512 header.b=NGMcPTTN; 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; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=foxmail.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id h7-20020aa7de07000000b00572ab8ed69esi2598900edv.541.2024.05.04.07.50.26; Sat, 04 May 2024 07:50:26 -0700 (PDT) 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=@foxmail.com header.s=s201512 header.b=NGMcPTTN; 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; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=foxmail.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id CA93E68D442; Sat, 4 May 2024 17:49:28 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from out162-62-58-211.mail.qq.com (out162-62-58-211.mail.qq.com [162.62.58.211]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id EFC6268D6BA for ; Sat, 4 May 2024 17:49:16 +0300 (EEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=foxmail.com; s=s201512; t=1714834149; bh=XXiVN8xaNKmFoohTBC5N1dOS/uRD9ljKqX7rLmP1utI=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=NGMcPTTNTDOhZSHP786ay7NYS18nwyh/JD7qjeIDMyPJX1H6P47Dv75UtlUoCtbsd g94fgtk6tjDV8U+s1/DGtWp+sIEMXuUIFXwYtHvm1+G8FZzd9suvpieGRESDvKWuNH SJt+oGZOtthKDvV+uL+yTaAzV2ExMgJ0QzkKylv0= Received: from localhost.localdomain ([42.56.223.122]) by newxmesmtplogicsvrsza29-0.qq.com (NewEsmtp) with SMTP id C411502C; Sat, 04 May 2024 22:49:01 +0800 X-QQ-mid: xmsmtpt1714834148tdnl55m2e Message-ID: X-QQ-XMAILINFO: OKKHiI6c9SH30kmgzvZhWw7qsO9Vx2wf9YTwr5qeWQh0rB1UcJRUyrL2OYp6Ws bM5SkWdnf9txWvcnVTvVy3mZ6PqGdax9Ua9UVBYPPdImUpcATWKUq+9PwZm9pU1T0FYdjsSEDxBE LggkEMPpfspCNHzWl67t3zBKlrzLLUhRkb0Q7SH4FnLRulxGUQtnYrTPv8bjrR60+InjVoF8owYE KmNYba+SN0mXQWVnT8Gcu0PM6Fdt962XHsdHl2fxWmAU+U8rVvDM4nBoU3dEuotUt9VfpfXhPV/1 34VSm8MtizorwKiIHOZP+50HI9qcuQkj5JT4aRdOB6o5LFNW351c4j/GjNl+IKgQRZ/slIsDJXbG klklLaSfabutyrU0pWka6l9zyZq784l1akIA0VxdJMfVZ/nLj1tMk7EqrEDMSt87/RwQNIji9zrL 6FLISUEMZlhhFNOFYak+U2JrLV2OkBQAfmiKjmwvD3AhalKuXAgsWeU97I664pYnldN9pXiTBqz/ 3UuN/dgGIPKvdzyK5FDa+7SVLy0bj7s0HD3T+nLR5MGoHGVAI4h49xYgI5406dBQDbscIRttCOkc yl8Or9ZD4yamzgqnjK5JiLgtaeCi23wk18gg2SZzXklOjsVm7esJ/IPaDIXl2zrnLX29Jxq9/IW9 eVLFy/zTTE2OQ1U7ikFE78pX0OqBMTUASNJyvlnb5DiL2i2gff5PnH2GuAxwQEY/Xjcflw++cskn bMas6QmNNo9aVUhH1412jwl1vkse8qdJvu5vzlmQXygK7421MmHozbh6TToH2qd/+QzyTrd4PKwu c8c7vk8vqYKf57AJJRWlhphjZH3xEU9pDUryO8SPekZv3NHJK3N0MvwQ7QB5aON/u61Anun7+Yn5 s5LI4gDLMQKiATI0NQkAKpYGDBLfy2fzo3zJ44EkRceUX0EKOexg+2KiIZhI5XrQ4Ap4SQI5UXyM 3ivw+TIbsfb8s51HrPtur2oH+Wz4sv X-QQ-XMRINFO: Mp0Kj//9VHAxr69bL5MkOOs= From: uk7b@foxmail.com To: ffmpeg-devel@ffmpeg.org Date: Sat, 4 May 2024 22:48:38 +0800 X-OQ-MSGID: <20240504144840.2411603-9-uk7b@foxmail.com> X-Mailer: git-send-email 2.45.0 In-Reply-To: <20240504144840.2411603-1-uk7b@foxmail.com> References: <20240504144840.2411603-1-uk7b@foxmail.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 09/10] lavc/vp8dsp: R-V V loop_filter_inner 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: sunyuechi Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: j6I12Jx1EmtF From: sunyuechi C908: vp8_loop_filter8uv_inner_v_c: 738.2 vp8_loop_filter8uv_inner_v_rvv_i32: 455.2 vp8_loop_filter16y_inner_h_c: 685.0 vp8_loop_filter16y_inner_h_rvv_i32: 497.0 vp8_loop_filter16y_inner_v_c: 743.7 vp8_loop_filter16y_inner_v_rvv_i32: 295.7 --- libavcodec/riscv/vp8dsp_init.c | 4 ++ libavcodec/riscv/vp8dsp_rvv.S | 110 +++++++++++++++++++++++++++++++++ 2 files changed, 114 insertions(+) diff --git a/libavcodec/riscv/vp8dsp_init.c b/libavcodec/riscv/vp8dsp_init.c index 46ca71ed04..aa95021df5 100644 --- a/libavcodec/riscv/vp8dsp_init.c +++ b/libavcodec/riscv/vp8dsp_init.c @@ -123,6 +123,10 @@ av_cold void ff_vp8dsp_init_riscv(VP8DSPContext *c) c->vp8_idct_dc_add4uv = ff_vp8_idct_dc_add4uv_rvv; } + c->vp8_v_loop_filter16y_inner = ff_vp8_v_loop_filter16_inner_rvv; + c->vp8_h_loop_filter16y_inner = ff_vp8_h_loop_filter16_inner_rvv; + c->vp8_v_loop_filter8uv_inner = ff_vp8_v_loop_filter8uv_inner_rvv; + c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter16_simple_rvv; c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter16_simple_rvv; } diff --git a/libavcodec/riscv/vp8dsp_rvv.S b/libavcodec/riscv/vp8dsp_rvv.S index 2eadfc5766..f10e269d9d 100644 --- a/libavcodec/riscv/vp8dsp_rvv.S +++ b/libavcodec/riscv/vp8dsp_rvv.S @@ -72,6 +72,13 @@ func ff_vp8_idct_dc_add4uv_rvv, zve32x ret endfunc +.macro filter_abs dst diff fI + vneg.v v8, \diff + vmax.vv \dst, v8, \diff + vmsleu.vx v8, \dst, \fI + vmand.mm v27, v27, v8 +.endm + .macro filter_fmin len a f1 p0f2 q0f1 .ifc \len,16 vsetvli zero, zero, e16, m2, ta, ma @@ -101,6 +108,16 @@ endfunc vle8.v v11, (t4) vle8.v v17, (t1) vle8.v v22, (\dst) + .if \normal + sub t3, t2, a6 + sub t0, t1, a6 + add t6, \dst, a6 + add a7, t4, a6 + vle8.v v2, (t3) + vle8.v v15, (t0) + vle8.v v10, (t6) + vle8.v v14, (a7) + .endif .else addi t1, \dst, -1 addi a6, \dst, -2 @@ -109,9 +126,28 @@ endfunc vlse8.v v11, (t4), \stride vlse8.v v17, (t1), \stride vlse8.v v22, (\dst), \stride + .if \normal + addi t5, \dst, -4 + addi t0, \dst, -3 + addi t6, \dst, 2 + addi a7, \dst, 3 + vlse8.v v2, (t5), \stride + vlse8.v v15, (t0), \stride + vlse8.v v10, (t6), \stride + vlse8.v v14, (a7), \stride + .endif .endif vwsubu.vv v12, v1, v11 // p1-q1 vwsubu.vv v24, v22, v17 // q0-p0 + +.if \normal + vwsubu.vv v30, v1, v17 + vwsubu.vv v20, v11, v22 + vwsubu.vv v28, v1, v15 + vwsubu.vv v4, v2, v15 + vwsubu.vv v6, v10, v11 + vwsubu.vv v2, v14, v10 +.endif vnclip.wi v23, v12, 0 .ifc \len,16 @@ -130,6 +166,26 @@ endfunc vmacc.vx v18, a7, v8 vmsleu.vx v0, v18, \fE +.if \normal + vneg.v v18, v30 + vmax.vv v30, v18, v30 + vmsleu.vx v27, v30, \fI + filter_abs v18 v28 \fI + filter_abs v18 v4 \fI + filter_abs v18 v6 \fI + filter_abs v18 v2 \fI + filter_abs v20 v20 \fI + vmand.mm v27, v0, v27 // vp8_simple_limit && normal + + vmsgtu.vx v20, v20, \thresh // hev + vmsgtu.vx v3, v30, \thresh + vmor.mm v3, v3, v20 // v3 = hev: > thresh + vzext.vf2 v18, v1 // v18 = p1 + vmand.mm v0, v27, v3 // v0 = normal && hev + vzext.vf2 v20, v11 // v12 = q1 + vmnot.m v3, v3 // v3 = !hv +.endif + li t5, 3 li a7, 124 li t3, 123 @@ -163,8 +219,62 @@ endfunc vsse8.v v6, (\dst), \stride, v0.t .endif +.if \normal + vmand.mm v0, v27, v3 // vp8_normal_limit & !hv + + .if \inner + vnclip.wi v30, v30, 0 + filter_fmin \len v30 v24 v4 v6 + vadd.vi v24, v24, 1 + vsra.vi v24, v24, 1 // (f1 + 1) >> 1; + vadd.vv v8, v18, v24 + vsub.vv v10, v20, v24 + .endif + + vmax.vx v8, v8, zero + vmax.vx v10, v10, zero +.ifc \len,16 + vsetvli zero, zero, e8, m1, ta, ma +.else + vsetvli zero, zero, e8, mf2, ta, ma +.endif + vnclipu.wi v4, v4, 0 + vnclipu.wi v5, v6, 0 + vnclipu.wi v6, v8, 0 + vnclipu.wi v7, v10, 0 + .ifc \type,v + vse8.v v4, (t1), v0.t + vse8.v v5, (\dst), v0.t + vse8.v v6, (t2), v0.t + vse8.v v7, (t4), v0.t + .else + vsse8.v v4, (t1), \stride, v0.t + vsse8.v v5, (\dst), \stride, v0.t + vsse8.v v6, (a6), \stride, v0.t + vsse8.v v7, (t4), \stride, v0.t + .endif +.endif .endm +func ff_vp8_h_loop_filter16_inner_rvv, zve32x + vsetivli zero, 16, e8, m1, ta, ma + filter 16 h 1 1 a0 a1 a2 a3 a4 + ret +endfunc + +func ff_vp8_v_loop_filter16_inner_rvv, zve32x + vsetivli zero, 16, e8, m1, ta, ma + filter 16 v 1 1 a0 a1 a2 a3 a4 + ret +endfunc + +func ff_vp8_v_loop_filter8uv_inner_rvv, zve32x + vsetivli zero, 8, e8, mf2, ta, ma + filter 8 v 1 1 a0 a2 a3 a4 a5 + filter 8 v 1 1 a1 a2 a3 a4 a5 + ret +endfunc + func ff_vp8_v_loop_filter16_simple_rvv, zve32x vsetivli zero, 16, e8, m1, ta, ma filter 16 v 0 0 a0 a1 a2 a3 a4