From patchwork Tue Jul 23 08:51:48 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: 50702 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a59:a742:0:b0:482:c625:d099 with SMTP id f2csp2486125vqm; Tue, 23 Jul 2024 01:54:39 -0700 (PDT) X-Forwarded-Encrypted: i=2; AJvYcCVHQx3KEHzlHODdkaaboj2c1D5USjqazwcQTd2EmVBNWwDAsABzOlOHuPWjM7XTjDllgOlb/me9e9G+mK0obBDBm3hLE1jFoNN9bg== X-Google-Smtp-Source: AGHT+IGVyi5OKYhOQbO8SOLshrlIsHOkD+oHQy8QiXr9sSZzPkka3+9+uofj8VoLZrG8TSx9Jx3u X-Received: by 2002:a2e:9b52:0:b0:2ef:248a:aed4 with SMTP id 38308e7fff4ca-2ef248ab33fmr56833901fa.42.1721724879589; Tue, 23 Jul 2024 01:54:39 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1721724879; cv=none; d=google.com; s=arc-20160816; b=RHfDhqS0d+etvna5OwK6pCxKZ12a22UVq3js1OKQfMqnhd0qc4JOcb86v/D0ZOhDL/ 8TQahYFYBTFtoC74R8r5gpSSRBXDfLd/VK9VbeymAH1rPTn5N6j8q3AO//W3NlAxL5ct vFmxrj7PrFXgadxEP2vznAafED7m6ZZvSax2W2nyCRG97TR4hGQ+CjHiINTY/Lb0hMgK Ayu6EjwdirsQZdoXG7yoztsQE3jEySc1PWSD26Micbs/eg0oU0GBz7/TrZgv/houNmDB k9XkE2+LyrV2131i9+syGrPctbJBhP9VtqkA3ktUZnsDkW74bnKRxG9JTb09XLlg0imz 7yDA== 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=AVZdRwtJPYJIkF5s4ZQ5PhboliY4JUcMVjmi1FBjsf8=; fh=D0bFwGkf4X22/D/bfeDVrXKIx7S6kcXsNzy10j8ORbQ=; b=J7sugWCu/yyL5XTn2xROdmACMMGdxaIaVVEO/gqpgVsi2x3a7hRRecKVLcwwWWCNpk WKsK4BPHphWcpK9X/FTX3Jri/Tshnzkjwh1gPtOZmSBUwTYsNjGYVgoJzQlgT7yYZOOR ySRKY3COEwaJMK0ExD1QGsxBMvRlLp7XUjqmpa3rm7XzZrpfXfQacWsSqSkIYqwU3Zdi ly0SorVKRvfOESl8jMaRO56MNpUQ4Civt65WNBnXHg7rbA/sC9rPmTCUE78JI53+EiY+ XrXAN0n2NjYgXauLL5OTTSWlC2RO2b+OUK6qts6Dyb3GUrCbLtG9zIu+MgecAUcAfW8L FK9g==; 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=zU8pCJpL; 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 38308e7fff4ca-2ef2553cd38si14737521fa.640.2024.07.23.01.54.39; Tue, 23 Jul 2024 01:54:39 -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=zU8pCJpL; 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 DBDAB68D588; Tue, 23 Jul 2024 11:54:33 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from xmbghk7.mail.qq.com (xmbghk7.mail.qq.com [43.163.128.49]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 4802068C872 for ; Tue, 23 Jul 2024 11:53:46 +0300 (EEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=foxmail.com; s=s201512; t=1721724714; bh=LBk2fPIo6aEEhAi6xu4tj5rtjyMCuu9IZ4jo+sFlpPo=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=zU8pCJpLFZ86gqcKnP81LAbX4ocKgVarInYH9q+K8xtHuEcZdZeEBFaV9ep7vTsib K2k0ClMuWZ1SPqIdzIKNd6grAZ0WXlqccmRTJbHMH3fXnhxOHB9PExyajOXn0q3/QV vViSTDMkLXoqsrSOSrb7gAbXpqDAaRHY7DIGOnNQ= Received: from localhost.localdomain ([42.86.132.58]) by newxmesmtplogicsvrszc13-0.qq.com (NewEsmtp) with SMTP id CF49A807; Tue, 23 Jul 2024 16:51:52 +0800 X-QQ-mid: xmsmtpt1721724712tspoy7zag Message-ID: X-QQ-XMAILINFO: NhpLzBn2I3XwjWnvg+I2QlMoOz/IMfN8704Zi/eoTexgoWKKnnhfvLR6vXq7Du oYGtxesK170tQcdQmr7TYbCt7WHpbc54VK6Pih6XbJcaCCOxd8ChDOxE1FZlWqiOFBA0i31zwgAa O+8Foo9kLTOpieVOQCOHkfPANA2PV53j28sOot7Wp4ScZXnQo7t39oQhluhVmz3guvpJ47U1avMs 5EkSxmdMQOIwJmUWFcV2bUR5BU07t61Do9VkVjZD/oDvraazAQz1kxLEf8Rd3DZQogE41kQpJGXE a58IPZ8rxYpQ9E/FoKcJIxyAGfGUtgpScVyZC6JUWt8Gy41t1L+3Df48uoltTfVqRSbmuKz8N9/u VY29iKwSaYTshOJNJUBSFBOvYvu2E9OjS9Krz4zjECKIGPE2SfhbmSoPwB9smeye7Hzl8u4VJS6+ 7BmYnY9UOy0onWdx7S7AXxOHpFtgSrYJ6yEg/H+nSvsbuS7bjawAQUCPBuVLQYTHRpInQHnK1C6U +3DzO4/uRB4v1Dg6Gs9Jvsmc8T1ogZx4bxkFNat7jYr5R8XZpZZJ1G7glW+RlfdmrHVR++81NZcZ BV98YOvb3T2sLjq1DDoLn1IgDrCUV5GN1I17Wuo0OW0t863he4X7zcpTJzrmHdcFpemda8QjJqkt m1MtAofLjG+KuN9fpI1Clta2R43VF8uLDzEhbqvPIUz0XJJksinkh3KPo0pTl77xUgatNneJ6mBZ EL1Fc48PR+eX5B70CFURja3bW63U+tzNBFI+L6T32Xr86YzEryvj2YkIuRpKN8gxVw6nyofOtwDs 8ToA0ZKclwlBIDPNWV3Ebgr232JtDGNn4zjsS1v+jZuqjGTr6QW4UXbhgul5ptr3PIIhc2RIJYjy JsnO78c6EQ1fW96rdSE3WFH12Hx6fg+g== X-QQ-XMRINFO: OWPUhxQsoeAVDbp3OJHYyFg= From: uk7b@foxmail.com To: ffmpeg-devel@ffmpeg.org Date: Tue, 23 Jul 2024 16:51:48 +0800 X-OQ-MSGID: <20240723085148.2639076-1-uk7b@foxmail.com> X-Mailer: git-send-email 2.45.2 In-Reply-To: <6069879.RfdxbOI5XL@basile.remlab.net> References: <6069879.RfdxbOI5XL@basile.remlab.net> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH v4 3/4] lavc/vp9dsp: R-V V mc tap h v 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: gGuSnFfLwUcZ From: sunyuechi C908 X60 vp9_avg_8tap_smooth_4h_8bpp_c : 12.7 11.2 vp9_avg_8tap_smooth_4h_8bpp_rvv_i32 : 4.7 4.2 vp9_avg_8tap_smooth_4v_8bpp_c : 29.7 12.5 vp9_avg_8tap_smooth_4v_8bpp_rvv_i32 : 4.7 4.2 vp9_avg_8tap_smooth_8h_8bpp_c : 48.7 42.2 vp9_avg_8tap_smooth_8h_8bpp_rvv_i32 : 9.5 8.5 vp9_avg_8tap_smooth_8v_8bpp_c : 49.7 45.5 vp9_avg_8tap_smooth_8v_8bpp_rvv_i32 : 9.5 8.5 vp9_avg_8tap_smooth_16h_8bpp_c : 192.0 166.5 vp9_avg_8tap_smooth_16h_8bpp_rvv_i32 : 21.7 19.5 vp9_avg_8tap_smooth_16v_8bpp_c : 191.2 175.2 vp9_avg_8tap_smooth_16v_8bpp_rvv_i32 : 21.2 19.0 vp9_avg_8tap_smooth_32h_8bpp_c : 780.2 663.2 vp9_avg_8tap_smooth_32h_8bpp_rvv_i32 : 68.2 60.5 vp9_avg_8tap_smooth_32v_8bpp_c : 770.0 685.7 vp9_avg_8tap_smooth_32v_8bpp_rvv_i32 : 67.0 59.5 vp9_avg_8tap_smooth_64h_8bpp_c : 3116.2 2648.2 vp9_avg_8tap_smooth_64h_8bpp_rvv_i32 : 270.7 120.7 vp9_avg_8tap_smooth_64v_8bpp_c : 3058.5 2731.7 vp9_avg_8tap_smooth_64v_8bpp_rvv_i32 : 266.5 119.0 vp9_put_8tap_smooth_4h_8bpp_c : 11.0 9.7 vp9_put_8tap_smooth_4h_8bpp_rvv_i32 : 4.2 3.7 vp9_put_8tap_smooth_4v_8bpp_c : 11.7 10.5 vp9_put_8tap_smooth_4v_8bpp_rvv_i32 : 4.0 3.7 vp9_put_8tap_smooth_8h_8bpp_c : 42.0 37.5 vp9_put_8tap_smooth_8h_8bpp_rvv_i32 : 8.5 7.7 vp9_put_8tap_smooth_8v_8bpp_c : 43.5 38.5 vp9_put_8tap_smooth_8v_8bpp_rvv_i32 : 8.7 7.7 vp9_put_8tap_smooth_16h_8bpp_c : 181.7 147.2 vp9_put_8tap_smooth_16h_8bpp_rvv_i32 : 20.0 18.0 vp9_put_8tap_smooth_16v_8bpp_c : 168.5 149.7 vp9_put_8tap_smooth_16v_8bpp_rvv_i32 : 19.7 17.5 vp9_put_8tap_smooth_32h_8bpp_c : 675.0 586.5 vp9_put_8tap_smooth_32h_8bpp_rvv_i32 : 65.2 58.0 vp9_put_8tap_smooth_32v_8bpp_c : 664.7 591.2 vp9_put_8tap_smooth_32v_8bpp_rvv_i32 : 64.0 57.0 vp9_put_8tap_smooth_64h_8bpp_c : 2696.2 2339.0 vp9_put_8tap_smooth_64h_8bpp_rvv_i32 : 259.7 115.7 vp9_put_8tap_smooth_64v_8bpp_c : 2691.0 2348.5 vp9_put_8tap_smooth_64v_8bpp_rvv_i32 : 255.5 114.0 --- libavcodec/riscv/vp9_mc_rvv.S | 193 +++++++++++++++++++++++++++++++++ libavcodec/riscv/vp9dsp.h | 72 ++++++++---- libavcodec/riscv/vp9dsp_init.c | 38 ++++++- 3 files changed, 278 insertions(+), 25 deletions(-) diff --git a/libavcodec/riscv/vp9_mc_rvv.S b/libavcodec/riscv/vp9_mc_rvv.S index 5241562531..6a4be7b9bd 100644 --- a/libavcodec/riscv/vp9_mc_rvv.S +++ b/libavcodec/riscv/vp9_mc_rvv.S @@ -36,6 +36,18 @@ .endif .endm +.macro vsetvlstatic16 len +.ifc \len,4 + vsetvli zero, zero, e16, mf2, ta, ma +.elseif \len == 8 + vsetvli zero, zero, e16, m1, ta, ma +.elseif \len == 16 + vsetvli zero, zero, e16, m2, ta, ma +.else + vsetvli zero, zero, e16, m4, ta, ma +.endif +.endm + .macro copy_avg len func ff_vp9_avg\len\()_rvv, zve32x csrwi vxrm, 0 @@ -181,8 +193,189 @@ func ff_\op\()_vp9_bilin_64hv_rvv, zve32x endfunc .endm +.equ ff_vp9_subpel_filters_smooth, ff_vp9_subpel_filters +.equ ff_vp9_subpel_filters_regular, ff_vp9_subpel_filters + 16*8*2 +.equ ff_vp9_subpel_filters_sharp, ff_vp9_subpel_filters + 16*8*2*2 + +.macro epel_filter name, type, regtype, arg + lla \regtype\()2, ff_vp9_subpel_filters_\name +.ifc \type,v + slli \regtype\()0, a6, 4 +.else + slli \regtype\()0, a5, 4 +.endif + add \regtype\()0, \regtype\()0, \regtype\()2 + lh \regtype\()1, 2(\regtype\()0) + lh \regtype\()2, 4(\regtype\()0) + lh \regtype\()3, 6(\regtype\()0) + lh \regtype\()4, 8(\regtype\()0) + lh \regtype\()5, 10(\regtype\()0) + lh \regtype\()6, 12(\regtype\()0) + lh \arg, 14(\regtype\()0) + lh \regtype\()0, 0(\regtype\()0) +.endm + +.macro epel_load dst, len, op, name, type, from_mem, regtype +.ifc \from_mem, 1 + vle8.v v22, (a2) +.ifc \type,v + add a5, a3, a2 + sub a2, a2, a3 + vle8.v v24, (a5) + vle8.v v20, (a2) + sh1add a2, a3, a5 + add a5, a5, a3 + vle8.v v26, (a5) + vle8.v v28, (a2) + add a2, a2, a3 + vle8.v v30, (a2) +.else + addi a5, a2, 1 + addi a2, a2, -1 + vle8.v v24, (a5) + vle8.v v20, (a2) + addi a5, a5, 2 + addi a2, a2, 3 + vle8.v v28, (a5) + vle8.v v26, (a2) + addi a2, a5, 1 + vle8.v v30, (a2) +.endif + +.ifc \name,smooth + vwmulu.vx v16, v24, \regtype\()4 + vwmaccu.vx v16, \regtype\()2, v20 + vwmaccu.vx v16, \regtype\()5, v26 + vwmaccsu.vx v16, \regtype\()6, v28 +.else + vwmulu.vx v16, v28, \regtype\()6 + vwmaccsu.vx v16, \regtype\()2, v20 + vwmaccsu.vx v16, \regtype\()5, v26 +.endif + +.ifc \regtype,t + vwmaccsu.vx v16, a7, v30 +.else + vwmaccsu.vx v16, s7, v30 +.endif + +.ifc \type,v + sh1add a5, a3, a3 + sub a2, a2, a5 + sub a2, a2, a5 + sub a5, a2, a3 + vle8.v v28, (a2) + vle8.v v26, (a5) + sh1add a2, a3, a2 +.else + addi a5, a2, -7 + addi a2, a2, -6 + vle8.v v26, (a5) + vle8.v v28, (a2) + addi a2, a2, 2 +.endif + +.ifc \name,smooth + vwmaccsu.vx v16, \regtype\()1, v28 +.else + vwmaccu.vx v16, \regtype\()1, v28 + vwmulu.vx v28, v24, \regtype\()4 +.endif + vwmaccsu.vx v16, \regtype\()0, v26 + vwmulu.vx v20, v22, \regtype\()3 +.else +.ifc \name,smooth + vwmulu.vx v16, v8, \regtype\()4 + vwmaccu.vx v16, \regtype\()2, v4 + vwmaccu.vx v16, \regtype\()5, v10 + vwmaccsu.vx v16, \regtype\()6, v12 + vwmaccsu.vx v16, \regtype\()1, v2 +.else + vwmulu.vx v16, v2, \regtype\()1 + vwmaccu.vx v16, \regtype\()6, v12 + vwmaccsu.vx v16, \regtype\()5, v10 + vwmaccsu.vx v16, \regtype\()2, v4 + vwmulu.vx v28, v8, \regtype\()4 +.endif + vwmaccsu.vx v16, \regtype\()0, v0 + vwmulu.vx v20, v6, \regtype\()3 + +.ifc \regtype,t + vwmaccsu.vx v16, a7, v14 +.else + vwmaccsu.vx v16, s7, v14 +.endif + +.endif + li a5, 64 + vwadd.wx v16, v16, a5 + vsetvlstatic16 \len + +.ifc \name,smooth + vwadd.vv v24, v16, v20 +.else + vwadd.vv v24, v16, v28 + vwadd.wv v24, v24, v20 +.endif + vnsra.wi v24, v24, 7 + vmax.vx v24, v24, zero + vsetvlstatic8 \len, zero, 32, m2 + + vnclipu.wi \dst, v24, 0 +.ifc \op,avg + vle8.v v24, (a0) + vaaddu.vv \dst, \dst, v24 +.endif + +.endm + +.macro epel_load_inc dst, len, op, name, type, from_mem, regtype + epel_load \dst, \len, \op, \name, \type, \from_mem, \regtype + add a2, a2, a3 +.endm + +.macro epel len, op, name, type, vlen +func ff_\op\()_vp9_8tap_\name\()_\len\()\type\()_rvv\vlen\(), zve32x + epel_filter \name, \type, t, a7 +.if \vlen < 256 + vsetvlstatic8 \len, a5, 32, m2 +.else + vsetvlstatic8 \len, a5, 64, m2 +.endif +.ifc \op,avg + csrwi vxrm, 0 +.endif + +1: + addi a4, a4, -1 + epel_load v30, \len, \op, \name, \type, 1, t + vse8.v v30, (a0) +.if \len == 64 && \vlen < 256 + addi a0, a0, 32 + addi a2, a2, 32 + epel_load v30, \len, \op, \name, \type, 1, t + vse8.v v30, (a0) + addi a0, a0, -32 + addi a2, a2, -32 +.endif + add a2, a2, a3 + add a0, a0, a1 + bnez a4, 1b + + ret +endfunc +.endm + .irp len, 64, 32, 16, 8, 4 copy_avg \len + .irp op, put, avg + .irp name, regular, sharp, smooth + .irp type, h, v + epel \len, \op, \name, \type, 128 + epel \len, \op, \name, \type, 256 + .endr + .endr + .endr .endr bilin_h_v put, h, a5 diff --git a/libavcodec/riscv/vp9dsp.h b/libavcodec/riscv/vp9dsp.h index 8fb326dae0..5fd64a1b8c 100644 --- a/libavcodec/riscv/vp9dsp.h +++ b/libavcodec/riscv/vp9dsp.h @@ -81,33 +81,39 @@ void ff_tm_8x8_rvv(uint8_t *dst, ptrdiff_t stride, const uint8_t *l, void ff_tm_4x4_rvv(uint8_t *dst, ptrdiff_t stride, const uint8_t *l, const uint8_t *a); -#define VP9_8TAP_RISCV_RVV_FUNC(SIZE, type, type_idx) \ -void ff_put_8tap_##type##_##SIZE##h_rvv(uint8_t *dst, ptrdiff_t dststride, \ +#define VP9_8TAP_RISCV_RVV_FUNC(SIZE, type, type_idx, min_vlen) \ +void ff_put_vp9_8tap_##type##_##SIZE##h_rvv##min_vlen(uint8_t *dst, \ + ptrdiff_t dststride, \ const uint8_t *src, \ ptrdiff_t srcstride, \ int h, int mx, int my); \ \ -void ff_put_8tap_##type##_##SIZE##v_rvv(uint8_t *dst, ptrdiff_t dststride, \ +void ff_put_vp9_8tap_##type##_##SIZE##v_rvv##min_vlen(uint8_t *dst, \ + ptrdiff_t dststride, \ const uint8_t *src, \ ptrdiff_t srcstride, \ int h, int mx, int my); \ \ -void ff_put_8tap_##type##_##SIZE##hv_rvv(uint8_t *dst, ptrdiff_t dststride, \ +void ff_put_vp9_8tap_##type##_##SIZE##hv_rvv##min_vlen(uint8_t *dst, \ + ptrdiff_t dststride, \ const uint8_t *src, \ ptrdiff_t srcstride, \ int h, int mx, int my); \ \ -void ff_avg_8tap_##type##_##SIZE##h_rvv(uint8_t *dst, ptrdiff_t dststride, \ +void ff_avg_vp9_8tap_##type##_##SIZE##h_rvv##min_vlen(uint8_t *dst, \ + ptrdiff_t dststride, \ const uint8_t *src, \ ptrdiff_t srcstride, \ int h, int mx, int my); \ \ -void ff_avg_8tap_##type##_##SIZE##v_rvv(uint8_t *dst, ptrdiff_t dststride, \ +void ff_avg_vp9_8tap_##type##_##SIZE##v_rvv##min_vlen(uint8_t *dst, \ + ptrdiff_t dststride, \ const uint8_t *src, \ ptrdiff_t srcstride, \ int h, int mx, int my); \ \ -void ff_avg_8tap_##type##_##SIZE##hv_rvv(uint8_t *dst, ptrdiff_t dststride, \ +void ff_avg_vp9_8tap_##type##_##SIZE##hv_rvv##min_vlen(uint8_t *dst, \ + ptrdiff_t dststride, \ const uint8_t *src, \ ptrdiff_t srcstride, \ int h, int mx, int my); @@ -146,23 +152,41 @@ void ff_vp9_avg##SIZE##_rvv(uint8_t *dst, ptrdiff_t dststride, \ const uint8_t *src, ptrdiff_t srcstride, \ int h, int mx, int my); -VP9_8TAP_RISCV_RVV_FUNC(64, regular, FILTER_8TAP_REGULAR); -VP9_8TAP_RISCV_RVV_FUNC(32, regular, FILTER_8TAP_REGULAR); -VP9_8TAP_RISCV_RVV_FUNC(16, regular, FILTER_8TAP_REGULAR); -VP9_8TAP_RISCV_RVV_FUNC(8, regular, FILTER_8TAP_REGULAR); -VP9_8TAP_RISCV_RVV_FUNC(4, regular, FILTER_8TAP_REGULAR); - -VP9_8TAP_RISCV_RVV_FUNC(64, sharp, FILTER_8TAP_SHARP); -VP9_8TAP_RISCV_RVV_FUNC(32, sharp, FILTER_8TAP_SHARP); -VP9_8TAP_RISCV_RVV_FUNC(16, sharp, FILTER_8TAP_SHARP); -VP9_8TAP_RISCV_RVV_FUNC(8, sharp, FILTER_8TAP_SHARP); -VP9_8TAP_RISCV_RVV_FUNC(4, sharp, FILTER_8TAP_SHARP); - -VP9_8TAP_RISCV_RVV_FUNC(64, smooth, FILTER_8TAP_SMOOTH); -VP9_8TAP_RISCV_RVV_FUNC(32, smooth, FILTER_8TAP_SMOOTH); -VP9_8TAP_RISCV_RVV_FUNC(16, smooth, FILTER_8TAP_SMOOTH); -VP9_8TAP_RISCV_RVV_FUNC(8, smooth, FILTER_8TAP_SMOOTH); -VP9_8TAP_RISCV_RVV_FUNC(4, smooth, FILTER_8TAP_SMOOTH); +VP9_8TAP_RISCV_RVV_FUNC(64, regular, FILTER_8TAP_REGULAR, 128); +VP9_8TAP_RISCV_RVV_FUNC(32, regular, FILTER_8TAP_REGULAR, 128); +VP9_8TAP_RISCV_RVV_FUNC(16, regular, FILTER_8TAP_REGULAR, 128); +VP9_8TAP_RISCV_RVV_FUNC(8, regular, FILTER_8TAP_REGULAR, 128); +VP9_8TAP_RISCV_RVV_FUNC(4, regular, FILTER_8TAP_REGULAR, 128); + +VP9_8TAP_RISCV_RVV_FUNC(64, sharp, FILTER_8TAP_SHARP, 128); +VP9_8TAP_RISCV_RVV_FUNC(32, sharp, FILTER_8TAP_SHARP, 128); +VP9_8TAP_RISCV_RVV_FUNC(16, sharp, FILTER_8TAP_SHARP, 128); +VP9_8TAP_RISCV_RVV_FUNC(8, sharp, FILTER_8TAP_SHARP, 128); +VP9_8TAP_RISCV_RVV_FUNC(4, sharp, FILTER_8TAP_SHARP, 128); + +VP9_8TAP_RISCV_RVV_FUNC(64, smooth, FILTER_8TAP_SMOOTH, 128); +VP9_8TAP_RISCV_RVV_FUNC(32, smooth, FILTER_8TAP_SMOOTH, 128); +VP9_8TAP_RISCV_RVV_FUNC(16, smooth, FILTER_8TAP_SMOOTH, 128); +VP9_8TAP_RISCV_RVV_FUNC(8, smooth, FILTER_8TAP_SMOOTH, 128); +VP9_8TAP_RISCV_RVV_FUNC(4, smooth, FILTER_8TAP_SMOOTH, 128); + +VP9_8TAP_RISCV_RVV_FUNC(64, regular, FILTER_8TAP_REGULAR, 256); +VP9_8TAP_RISCV_RVV_FUNC(32, regular, FILTER_8TAP_REGULAR, 256); +VP9_8TAP_RISCV_RVV_FUNC(16, regular, FILTER_8TAP_REGULAR, 256); +VP9_8TAP_RISCV_RVV_FUNC(8, regular, FILTER_8TAP_REGULAR, 256); +VP9_8TAP_RISCV_RVV_FUNC(4, regular, FILTER_8TAP_REGULAR, 256); + +VP9_8TAP_RISCV_RVV_FUNC(64, sharp, FILTER_8TAP_SHARP, 256); +VP9_8TAP_RISCV_RVV_FUNC(32, sharp, FILTER_8TAP_SHARP, 256); +VP9_8TAP_RISCV_RVV_FUNC(16, sharp, FILTER_8TAP_SHARP, 256); +VP9_8TAP_RISCV_RVV_FUNC(8, sharp, FILTER_8TAP_SHARP, 256); +VP9_8TAP_RISCV_RVV_FUNC(4, sharp, FILTER_8TAP_SHARP, 256); + +VP9_8TAP_RISCV_RVV_FUNC(64, smooth, FILTER_8TAP_SMOOTH, 256); +VP9_8TAP_RISCV_RVV_FUNC(32, smooth, FILTER_8TAP_SMOOTH, 256); +VP9_8TAP_RISCV_RVV_FUNC(16, smooth, FILTER_8TAP_SMOOTH, 256); +VP9_8TAP_RISCV_RVV_FUNC(8, smooth, FILTER_8TAP_SMOOTH, 256); +VP9_8TAP_RISCV_RVV_FUNC(4, smooth, FILTER_8TAP_SMOOTH, 256); VP9_BILINEAR_RISCV_RVV_FUNC(64); VP9_BILINEAR_RISCV_RVV_FUNC(32); diff --git a/libavcodec/riscv/vp9dsp_init.c b/libavcodec/riscv/vp9dsp_init.c index b3700dfb08..3669070fca 100644 --- a/libavcodec/riscv/vp9dsp_init.c +++ b/libavcodec/riscv/vp9dsp_init.c @@ -49,7 +49,9 @@ static av_cold void vp9dsp_mc_init_riscv(VP9DSPContext *dsp, int bpp) # endif #if HAVE_RVV - if (bpp == 8 && (flags & AV_CPU_FLAG_RVV_I32) && ff_rv_vlen_least(128)) { + if (bpp == 8 && (flags & AV_CPU_FLAG_RVV_I32)) { + int vlenb = ff_get_rv_vlenb(); + if (vlenb >= 16) { #define init_fpel(idx1, sz) \ dsp->mc[idx1][FILTER_8TAP_SMOOTH ][1][0][0] = ff_vp9_avg##sz##_rvv; \ @@ -95,6 +97,40 @@ static av_cold void vp9dsp_mc_init_riscv(VP9DSPContext *dsp, int bpp) dsp->mc[4][FILTER_BILINEAR ][1][1][1] = ff_avg_vp9_bilin_4hv_rvv; #undef init_fpel + +#define init_subpel1(idx1, idx2, idxh, idxv, sz, dir, type, vlen) \ + dsp->mc[idx1][FILTER_8TAP_SMOOTH ][idx2][idxh][idxv] = \ + ff_##type##_vp9_8tap_smooth_##sz##dir##_rvv##vlen; \ + dsp->mc[idx1][FILTER_8TAP_REGULAR][idx2][idxh][idxv] = \ + ff_##type##_vp9_8tap_regular_##sz##dir##_rvv##vlen; \ + dsp->mc[idx1][FILTER_8TAP_SHARP ][idx2][idxh][idxv] = \ + ff_##type##_vp9_8tap_sharp_##sz##dir##_rvv##vlen; + +#define init_subpel2(idx, idxh, idxv, dir, type, vlen) \ + init_subpel1(0, idx, idxh, idxv, 64, dir, type, vlen); \ + init_subpel1(1, idx, idxh, idxv, 32, dir, type, vlen); \ + init_subpel1(2, idx, idxh, idxv, 16, dir, type, vlen); \ + init_subpel1(3, idx, idxh, idxv, 8, dir, type, vlen); \ + init_subpel1(4, idx, idxh, idxv, 4, dir, type, vlen) + + init_subpel2(0, 1, 0, h, put, 128); + init_subpel2(1, 1, 0, h, avg, 128); + + if (flags & AV_CPU_FLAG_RVB_ADDR) { + init_subpel2(0, 0, 1, v, put, 128); + init_subpel2(1, 0, 1, v, avg, 128); + } + + } + if (vlenb >= 32) { + init_subpel2(0, 1, 0, h, put, 256); + init_subpel2(1, 1, 0, h, avg, 256); + + if (flags & AV_CPU_FLAG_RVB_ADDR) { + init_subpel2(0, 0, 1, v, put, 256); + init_subpel2(1, 0, 1, v, avg, 256); + } + } } #endif #endif From patchwork Tue Jul 23 08:58:13 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: 50703 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a59:a742:0:b0:482:c625:d099 with SMTP id f2csp2487774vqm; Tue, 23 Jul 2024 01:59:12 -0700 (PDT) X-Forwarded-Encrypted: i=2; AJvYcCWeXHI99ZIQaUr2hyEzUiyTYDMxMFIs3EmV/kI2V1HcDaEGIUjuMX3fFj+3zAebKzhA5vNMtfxrEWaeWYjoa4gniYPksPLPxJdzHQ== X-Google-Smtp-Source: AGHT+IEM7dAvgFv9xU5NyjZoIL3hi/xDGMjaiFcfNfadrAJeBJNed6E47rdO7Zj/Hxw0Zr/31V7b X-Received: by 2002:a50:d75b:0:b0:5a3:76e3:1dc with SMTP id 4fb4d7f45d1cf-5a479d6279bmr5158249a12.10.1721725152472; Tue, 23 Jul 2024 01:59:12 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1721725152; cv=none; d=google.com; s=arc-20160816; b=tMM8KXRDlEdgomkU4KmpJdhRALiLKzxTLHkPpIZBQIeBN31uHAP6aiomBMP09rTSQ/ qOVsX94LY7k3wJ1jOECohu/v1REhjcGwHm0oVJTeLaS9d7jvGNv61zST3KOlKFv4FZW7 q7DbM4B3k1ywAxZzCOtVufFOLIo4bUWXOemiGLZ/kk0eX69UVoW6luhzUsoOGMszH1JW u8CCFwkFtl0/5AvVy6KOq3IVIBm2FVsCHnYEsoesK9n0Dq/+fuJ49gjnfAP1he0RsT/9 ISnOdLtHAv/JrMldGQd9x1DzkJE+6DZ6X6ESBaRaT9r119ABMp+Ks3+YTcLmeP5u8D0F G5KA== 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=lVvUAyfpm/P5ftLZuQrwuv9Nxyc/78JJ1BwwfHV1S8I=; fh=D0bFwGkf4X22/D/bfeDVrXKIx7S6kcXsNzy10j8ORbQ=; b=U2GqWNop5kB1twHlPhYQyXWe2LGLvpJdtaVtSWiaRnWrKsGZoUcEwTeMkRi+LrHH3D ofcFxERsQ24krC0P5/seNQdC/uLgNLvzUGT2vFZ7/eDnp31GBVuIapaPoEjx+jZCkXkN sqha9v54th8/pXzFFuFaV6zn7v4OFeS6YARbHplNoOav9pr0OFc3AtJxxsq0G5W1TcuK qyKHCAdqV2FTjbBsoti2SJr3i5aZnM8VHR0ZyXOb1oHN1/UclJck4mRlxuh7Gr2CSzCW 34U0tPj6DeEVJBiOQmZ6DmfI5Y07DEDXNKjAxWN5Q8obp+YOBfouJz9FZ5fSR33Ye210 /4IA==; 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=YPCQTVeN; 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 4fb4d7f45d1cf-5a30a0006b2si5778950a12.126.2024.07.23.01.59.12; Tue, 23 Jul 2024 01:59:12 -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=YPCQTVeN; 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 3EBE168D45F; Tue, 23 Jul 2024 11:59:10 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from out162-62-57-210.mail.qq.com (out162-62-57-210.mail.qq.com [162.62.57.210]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 456F668D45F for ; Tue, 23 Jul 2024 11:59:01 +0300 (EEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=foxmail.com; s=s201512; t=1721725096; bh=8y5l4fYqstbPilGVzJi0ZhFNtbBeI/kUBA9jIRr/cg4=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=YPCQTVeN8yNlrabP7lo42a3ZBH/0q3E6QNV8y0vUQ4tK1sdbk7ongUvdq4nf9MHsf dkKCRwxydiX8Uh0v7XNBvmJ1WxI9A/zUYTVHlmqDAs5DtNds1IyuMPuEMmIa8a5fee BX2yN1gs44Vc0lemZm2GZ52fKrbyfxGdlzmjfVlU= Received: from localhost.localdomain ([42.86.132.58]) by newxmesmtplogicsvrsza29-0.qq.com (NewEsmtp) with SMTP id E8F1EAEF; Tue, 23 Jul 2024 16:58:15 +0800 X-QQ-mid: xmsmtpt1721725095tz3w068nc Message-ID: X-QQ-XMAILINFO: OOPJ7pYMv25ttAe9j/0hzrTaFcBBtFeLnbs+krB3oeUzdRYX39tpbvtxhL+viR mBpPxCjkj9iUeo+Dr98PiyPlOgXishVyLrKd2p03bq+nP2Srfq0OTIB6qlXs6Mr6aaJ8cjOSuQqt orph/1YJfLGq3wXY03suCzohbFWla4XOABOYl3UH+Uxnq9FieZAE9+k8IQxVH4F0rW4XXSRKJBY9 /XOZDFltzn9MQCg7WJqyr2VCR110WvNeh+7H2IeyLTGQRl7Zldtd7OiefgAAff/1tG3LAny2qVCg Ju9tkkdhrVwMK1EQ536irnXlEa6//MbXj7jp13U2UJyj/PpoccIrqPmR8UTVzExGtJNRvAmxJHwO CV9cppJnYSHMpAJeUwFR0baNNYCdWYJ0kAzquBYpauV84vEl8Knzm8RpQd36Di1V47tJKVN+vOW2 q/q5xDVvsWTGDXK1+abmNWuP8lZjnhZDzSTTWSyMtbaVuehCaKdIg7avQfA+FC6j7vgN9YIfxIzR lhPPkCK8dbmZGa3FRIZq+CUxawcsUFjxO7Hcj6mL1dc6tbl0niEQHfQz0s/hxruKUo6KzSLIetI0 MGgE0nDX+vcyK33KEJ1Dx1cIsyqxpAyx3cLGs2Vp8IhcD7lq502vpye5OnaOj6FWsIXYB+lQbRbU eOHJCDPUGbZI1BP92/qcDlmyxX9VKwsFujxt7hwn7oIK9K0//5yfn/8iEATYEAJ/lLJ0wta9W4Ou AVdIbbfH9Vf8clU3cK6CD1ox7Agd8cLYrDDecqrzquzG9L7u58Ssx4XfieliGPjECHe1gGY1vgeM lLPH2/jDLcu8kFcJRnn6AOL4VEdl92qFi/Zc3wpOSzRc3XY9/3EINL7VRCm//2x1hWC0+6tRlp/z MVF7/3ySRbUBO/cEjpV9Fnm/AYeVQIdL3oktDzpAuL X-QQ-XMRINFO: M/715EihBoGSf6IYSX1iLFg= From: uk7b@foxmail.com To: ffmpeg-devel@ffmpeg.org Date: Tue, 23 Jul 2024 16:58:13 +0800 X-OQ-MSGID: <20240723085813.2683769-1-uk7b@foxmail.com> X-Mailer: git-send-email 2.45.2 In-Reply-To: References: MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH v4 4/4] lavc/vp9dsp: R-V V mc tap hv 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: PU7bDYddQhsK From: sunyuechi C908 X60 vp9_avg_8tap_smooth_4hv_8bpp_c : 32.0 28.0 vp9_avg_8tap_smooth_4hv_8bpp_rvv_i32 : 15.0 13.2 vp9_avg_8tap_smooth_8hv_8bpp_c : 98.0 86.2 vp9_avg_8tap_smooth_8hv_8bpp_rvv_i32 : 23.7 21.2 vp9_avg_8tap_smooth_16hv_8bpp_c : 355.7 297.0 vp9_avg_8tap_smooth_16hv_8bpp_rvv_i32 : 47.0 41.5 vp9_avg_8tap_smooth_32hv_8bpp_c : 1272.7 1099.7 vp9_avg_8tap_smooth_32hv_8bpp_rvv_i32 : 134.7 119.7 vp9_avg_8tap_smooth_64hv_8bpp_c : 4937.0 4224.2 vp9_avg_8tap_smooth_64hv_8bpp_rvv_i32 : 528.5 228.5 vp9_put_8tap_smooth_4hv_8bpp_c : 30.2 26.7 vp9_put_8tap_smooth_4hv_8bpp_rvv_i32 : 30.5 12.5 vp9_put_8tap_smooth_8hv_8bpp_c : 91.5 81.2 vp9_put_8tap_smooth_8hv_8bpp_rvv_i32 : 22.7 20.2 vp9_put_8tap_smooth_16hv_8bpp_c : 313.2 277.5 vp9_put_8tap_smooth_16hv_8bpp_rvv_i32 : 45.2 40.2 vp9_put_8tap_smooth_32hv_8bpp_c : 1166.7 1022.2 vp9_put_8tap_smooth_32hv_8bpp_rvv_i32 : 131.7 117.2 vp9_put_8tap_smooth_64hv_8bpp_c : 4560.5 3961.7 vp9_put_8tap_smooth_64hv_8bpp_rvv_i32 : 517.0 223.2 --- libavcodec/riscv/vp9_mc_rvv.S | 75 ++++++++++++++++++++++++++++++++++ libavcodec/riscv/vp9dsp_init.c | 8 ++++ 2 files changed, 83 insertions(+) diff --git a/libavcodec/riscv/vp9_mc_rvv.S b/libavcodec/riscv/vp9_mc_rvv.S index 6a4be7b9bd..26754ac6f8 100644 --- a/libavcodec/riscv/vp9_mc_rvv.S +++ b/libavcodec/riscv/vp9_mc_rvv.S @@ -366,6 +366,77 @@ func ff_\op\()_vp9_8tap_\name\()_\len\()\type\()_rvv\vlen\(), zve32x endfunc .endm +#if __riscv_xlen == 64 +.macro epel_hv_once len, name, op + sub a2, a2, a3 + sub a2, a2, a3 + sub a2, a2, a3 + .irp n,0,2,4,6,8,10,12,14 + epel_load_inc v\n, \len, put, \name, h, 1, t + .endr + addi a4, a4, -1 +1: + addi a4, a4, -1 + epel_load v30, \len, \op, \name, v, 0, s + vse8.v v30, (a0) + vmv.v.v v0, v2 + vmv.v.v v2, v4 + vmv.v.v v4, v6 + vmv.v.v v6, v8 + vmv.v.v v8, v10 + vmv.v.v v10, v12 + vmv.v.v v12, v14 + epel_load v14, \len, put, \name, h, 1, t + add a2, a2, a3 + add a0, a0, a1 + bnez a4, 1b + epel_load v30, \len, \op, \name, v, 0, s + vse8.v v30, (a0) +.endm + +.macro epel_hv op, name, len, vlen +func ff_\op\()_vp9_8tap_\name\()_\len\()hv_rvv\vlen\(), zve32x + addi sp, sp, -64 + .irp n,0,1,2,3,4,5,6,7 + sd s\n, \n\()<<3(sp) + .endr +.if \len == 64 && \vlen < 256 + addi sp, sp, -48 + .irp n,0,1,2,3,4,5 + sd a\n, \n\()<<3(sp) + .endr +.endif +.ifc \op,avg + csrwi vxrm, 0 +.endif + epel_filter \name, h, t, a7 + epel_filter \name, v, s, s7 +.if \vlen < 256 + vsetvlstatic8 \len, a6, 32, m2 +.else + vsetvlstatic8 \len, a6, 64, m2 +.endif + epel_hv_once \len, \name, \op +.if \len == 64 && \vlen < 256 + .irp n,0,1,2,3,4,5 + ld a\n, \n\()<<3(sp) + .endr + addi sp, sp, 48 + addi a0, a0, 32 + addi a2, a2, 32 + epel_filter \name, h, t, a7 + epel_hv_once \len, \name, \op +.endif + .irp n,0,1,2,3,4,5,6,7 + ld s\n, \n\()<<3(sp) + .endr + addi sp, sp, 64 + + ret +endfunc +.endm +#endif + .irp len, 64, 32, 16, 8, 4 copy_avg \len .irp op, put, avg @@ -374,6 +445,10 @@ endfunc epel \len, \op, \name, \type, 128 epel \len, \op, \name, \type, 256 .endr + #if __riscv_xlen == 64 + epel_hv \op, \name, \len, 128 + epel_hv \op, \name, \len, 256 + #endif .endr .endr .endr diff --git a/libavcodec/riscv/vp9dsp_init.c b/libavcodec/riscv/vp9dsp_init.c index 3669070fca..7b090c9889 100644 --- a/libavcodec/riscv/vp9dsp_init.c +++ b/libavcodec/riscv/vp9dsp_init.c @@ -119,6 +119,10 @@ static av_cold void vp9dsp_mc_init_riscv(VP9DSPContext *dsp, int bpp) if (flags & AV_CPU_FLAG_RVB_ADDR) { init_subpel2(0, 0, 1, v, put, 128); init_subpel2(1, 0, 1, v, avg, 128); +# if __riscv_xlen == 64 + init_subpel2(0, 1, 1, hv, put, 128); + init_subpel2(1, 1, 1, hv, avg, 128); +# endif } } @@ -129,6 +133,10 @@ static av_cold void vp9dsp_mc_init_riscv(VP9DSPContext *dsp, int bpp) if (flags & AV_CPU_FLAG_RVB_ADDR) { init_subpel2(0, 0, 1, v, put, 256); init_subpel2(1, 0, 1, v, avg, 256); +# if __riscv_xlen == 64 + init_subpel2(0, 1, 1, hv, put, 256); + init_subpel2(1, 1, 1, hv, avg, 256); +# endif } } }