From patchwork Thu Jun 6 18:26:15 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= X-Patchwork-Id: 49628 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a59:c209:0:b0:460:55fa:d5ed with SMTP id d9csp564627vqo; Thu, 6 Jun 2024 11:26:28 -0700 (PDT) X-Forwarded-Encrypted: i=2; AJvYcCVgk03Zfxcn4tXUTDHiJ9REJ8HQrhx/Zvb0RpzXpjycR7I4RFN8QwvMCUZ/O15fl1N9lCGmLDLXXywC5nT73iPcQmnmLq4ql7KpoQ== X-Google-Smtp-Source: AGHT+IHnkC3Izyvm7pVO9mTe7HSP6Hmg/JUObEUik3DBrbmMGra5avz/A2oVl9kYEMzz6smIZ/gx X-Received: by 2002:a17:906:aec4:b0:a6c:84e3:fda3 with SMTP id a640c23a62f3a-a6cdb9dc88fmr25220366b.5.1717698388490; Thu, 06 Jun 2024 11:26:28 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1717698388; cv=none; d=google.com; s=arc-20160816; b=mDh2fsonQ22kg8MxXy0sk7vphlBM3TLj53ffmWQo+ZzKuBJIfa7awt96tiuch+xZpd UZp+/MxUAdwhvaCsuc7Bp2N/Nr1G1BQ+53m71xwj0sYkIs9qa9QKY0AsI8fEf08VEE1z I4YwfEO4MybSVsbcL9Idc5aLiMI2j0znATm/eDbFk3qTXcqx86zCSAmXnEoQ6cEEqZ7v 8FnCjyvze8v+TBvqrxuHcgsbc30V7YqoWrLnza4VxKua537wAFQZ2pcdiGxRPWRVICoI PHcVi35rw/VtnowtwJ7MiLYrOWCsplX5A+FfJ22JjJ3baN8EWVXRtRBwVb/5MTAFcwza 46bA== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:reply-to:list-subscribe :list-help:list-post:list-archive:list-unsubscribe:list-id :precedence:subject:mime-version:message-id:date:to:from :delivered-to; bh=Tocf1Mzimx5I3NJ6zRNp+ABNiHtzVGa+BAhT58UrFSo=; fh=YOA8vD9MJZuwZ71F/05pj6KdCjf6jQRmzLS+CATXUQk=; b=qdv0HqRzGYYKl9pO0ZOuZEBC3UYiQa3Pl0O2ZxTgJghc6kArEOAIf5IG1vE3LzgVAi G8ArV4/+U8RnaeNnsyLstoUbvn3JuTgo/viAxfOU0Y1H6r1/iBdxMY3D1LPWKXfYFHt8 bnQWQD6JZtVcqtC342gNfHatVbdoG4a3jueqsHAiWLpE/3EY8PCM9/ikScnfr+M7rgyp cTSXJ0HoghQrbXMJ3lTUiVbr8DlGdWwpNd+ZCijn2zEaRbCPWGXdQtGaqf93yFsqz15o 7m6d20Qlg9T8NF7g0czf29lm3Ubm5GHb1+xCFAmM0Zal7Z4RJykOllpdbKU88CCUtav4 zSOQ==; dara=google.com ARC-Authentication-Results: i=1; mx.google.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 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id a640c23a62f3a-a6c806dad83si91945966b.383.2024.06.06.11.26.28; Thu, 06 Jun 2024 11:26:28 -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; 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 D236B68D6E2; Thu, 6 Jun 2024 21:26:25 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from ursule.remlab.net (vps-a2bccee9.vps.ovh.net [51.75.19.47]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 77F0068D6E2 for ; Thu, 6 Jun 2024 21:26:18 +0300 (EEST) Received: from basile.remlab.net (localhost [IPv6:::1]) by ursule.remlab.net (Postfix) with ESMTP id ECE22C0090 for ; Thu, 6 Jun 2024 21:26:17 +0300 (EEST) From: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= To: ffmpeg-devel@ffmpeg.org Date: Thu, 6 Jun 2024 21:26:15 +0300 Message-ID: <20240606182617.39636-1-remi@remlab.net> X-Mailer: git-send-email 2.45.1 MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 1/3] sws/input: R-V V 32-bit RGB to Y 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 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: DzCaLdJmqm+h T-Head C908: abgr_to_y_8_c: 2.5 abgr_to_y_8_rvv_i32: 2.2 abgr_to_y_128_c: 37.0 abgr_to_y_128_rvv_i32: 8.5 abgr_to_y_1080_c: 327.0 abgr_to_y_1080_rvv_i32: 69.5 abgr_to_y_1920_c: 552.0 abgr_to_y_1920_rvv_i32: 122.2 bgra_to_y_8_c: 2.5 bgra_to_y_8_rvv_i32: 2.2 bgra_to_y_128_c: 37.2 bgra_to_y_128_rvv_i32: 8.5 bgra_to_y_1080_c: 310.2 bgra_to_y_1080_rvv_i32: 69.5 bgra_to_y_1920_c: 568.2 bgra_to_y_1920_rvv_i32: 122.5 SpacemiT X60: abgr_to_y_8_c: 2.5 abgr_to_y_8_rvv_i32: 2.0 abgr_to_y_128_c: 33.0 abgr_to_y_128_rvv_i32: 3.7 abgr_to_y_1080_c: 276.0 abgr_to_y_1080_rvv_i32: 31.5 abgr_to_y_1920_c: 493.7 abgr_to_y_1920_rvv_i32: 55.5 bgra_to_y_8_c: 2.2 bgra_to_y_8_rvv_i32: 2.0 bgra_to_y_128_c: 33.0 bgra_to_y_128_rvv_i32: 3.7 bgra_to_y_1080_c: 276.0 bgra_to_y_1080_rvv_i32: 31.5 bgra_to_y_1920_c: 490.7 bgra_to_y_1920_rvv_i32: 55.5 --- libswscale/riscv/input.S | 45 +++++++++++++++++++++++++++++++++++++ libswscale/riscv/swscale.c | 46 ++++++++++++++++++++++++++------------ 2 files changed, 77 insertions(+), 14 deletions(-) diff --git a/libswscale/riscv/input.S b/libswscale/riscv/input.S index f84f79899b..8dfd012a9d 100644 --- a/libswscale/riscv/input.S +++ b/libswscale/riscv/input.S @@ -149,3 +149,48 @@ func ff_rgb24ToUV_half_rvv, zve32x ret endfunc + + .macro rgba_input chr0, chr1, high +func ff_\chr1\()ToY_rvv, zve32x + lw t1, 8(a5) # BY + lw t3, 0(a5) # RY + j 1f +endfunc + +func ff_\chr0\()ToY_rvv, zve32x + lw t1, 0(a5) # RY + lw t3, 8(a5) # BY +1: + lw t2, 4(a5) # GY + li t4, (32 << (15 - 1)) + (1 << (15 - 7)) + li t5, 0xff +2: + vsetvli t0, a4, e32, m8, ta, ma + vle32.v v0, (a1) + sub a4, a4, t0 + .if \high + vsrl.vi v8, v0, 24 + .else + vand.vx v8, v0, t5 + .endif + sh2add a1, t0, a1 + vsrl.vi v16, v0, 8 * (1 + \high) + vmul.vx v24, v8, t1 + vand.vx v16, v16, t5 + vsrl.vi v8, v0, 8 * (2 - \high) + vmacc.vx v24, t2, v16 + vand.vx v8, v8, t5 + vadd.vx v24, v24, t4 + vmacc.vx v24, t3, v8 + vsetvli zero, zero, e16, m4, ta, ma + vnsra.wi v0, v24, 15 - 6 + vse16.v v0, (a0) + sh1add a0, t0, a0 + bnez a4, 2b + + ret +endfunc + .endm + +rgba_input rgba32, bgra32, 0 +rgba_input abgr32, argb32, 1 diff --git a/libswscale/riscv/swscale.c b/libswscale/riscv/swscale.c index bae626f64a..fcb338b295 100644 --- a/libswscale/riscv/swscale.c +++ b/libswscale/riscv/swscale.c @@ -21,20 +21,22 @@ #include "libavutil/riscv/cpu.h" #include "libswscale/swscale_internal.h" -void ff_bgr24ToY_rvv(uint8_t *dst, const uint8_t *src, const uint8_t *, - const uint8_t *, int width, uint32_t *coeffs, void *); -void ff_bgr24ToUV_rvv(uint8_t *, uint8_t *, const uint8_t *, const uint8_t *, - const uint8_t *, int width, uint32_t *coeffs, void *); -void ff_bgr24ToUV_half_rvv(uint8_t *, uint8_t *, const uint8_t *, - const uint8_t *, const uint8_t *, int width, - uint32_t *coeffs, void *); -void ff_rgb24ToY_rvv(uint8_t *dst, const uint8_t *src, const uint8_t *, - const uint8_t *, int width, uint32_t *coeffs, void *); -void ff_rgb24ToUV_rvv(uint8_t *, uint8_t *, const uint8_t *, const uint8_t *, - const uint8_t *, int width, uint32_t *coeffs, void *); -void ff_rgb24ToUV_half_rvv(uint8_t *, uint8_t *, const uint8_t *, - const uint8_t *, const uint8_t *, int width, - uint32_t *coeffs, void *); +#define RVV_INPUT(name) \ +void ff_##name##ToY_rvv(uint8_t *dst, const uint8_t *src, const uint8_t *, \ + const uint8_t *, int w, uint32_t *coeffs, void *); \ +void ff_##name##ToUV_rvv(uint8_t *, uint8_t *, const uint8_t *, \ + const uint8_t *, const uint8_t *, int w, \ + uint32_t *coeffs, void *); \ +void ff_##name##ToUV_half_rvv(uint8_t *, uint8_t *, const uint8_t *, \ + const uint8_t *, const uint8_t *, int w, \ + uint32_t *coeffs, void *) + +RVV_INPUT(abgr32); +RVV_INPUT(argb32); +RVV_INPUT(bgr24); +RVV_INPUT(bgra32); +RVV_INPUT(rgb24); +RVV_INPUT(rgba32); av_cold void ff_sws_init_swscale_riscv(SwsContext *c) { @@ -43,6 +45,14 @@ av_cold void ff_sws_init_swscale_riscv(SwsContext *c) if ((flags & AV_CPU_FLAG_RVV_I32) && (flags & AV_CPU_FLAG_RVB_ADDR)) { switch (c->srcFormat) { + case AV_PIX_FMT_ABGR: + c->lumToYV12 = ff_abgr32ToY_rvv; + break; + + case AV_PIX_FMT_ARGB: + c->lumToYV12 = ff_argb32ToY_rvv; + break; + case AV_PIX_FMT_BGR24: c->lumToYV12 = ff_bgr24ToY_rvv; if (c->chrSrcHSubSample) @@ -51,6 +61,10 @@ av_cold void ff_sws_init_swscale_riscv(SwsContext *c) c->chrToYV12 = ff_bgr24ToUV_rvv; break; + case AV_PIX_FMT_BGRA: + c->lumToYV12 = ff_bgra32ToY_rvv; + break; + case AV_PIX_FMT_RGB24: c->lumToYV12 = ff_rgb24ToY_rvv; if (c->chrSrcHSubSample) @@ -58,6 +72,10 @@ av_cold void ff_sws_init_swscale_riscv(SwsContext *c) else c->chrToYV12 = ff_rgb24ToUV_rvv; break; + + case AV_PIX_FMT_RGBA: + c->lumToYV12 = ff_rgba32ToY_rvv; + break; } } #endif