From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org [79.124.17.100]) by master.gitmailbox.com (Postfix) with ESMTP id CF1E14B444 for ; Wed, 5 Jun 2024 16:36:56 +0000 (UTC) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id B53DC68D6FE; Wed, 5 Jun 2024 19:36:46 +0300 (EEST) Received: from ursule.remlab.net (vps-a2bccee9.vps.ovh.net [51.75.19.47]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id B4BA768D67D for ; Wed, 5 Jun 2024 19:36:39 +0300 (EEST) Received: from basile.remlab.net (localhost [IPv6:::1]) by ursule.remlab.net (Postfix) with ESMTP id 527F2C02EF for ; Wed, 5 Jun 2024 19:36:39 +0300 (EEST) From: =?UTF-8?q?R=C3=A9mi=20Denis-Courmont?= To: ffmpeg-devel@ffmpeg.org Date: Wed, 5 Jun 2024 19:36:36 +0300 Message-ID: <20240605163638.10586-2-remi@remlab.net> X-Mailer: git-send-email 2.45.1 In-Reply-To: <20240605163638.10586-1-remi@remlab.net> References: <20240605163638.10586-1-remi@remlab.net> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 2/3] sws/input: R-V V rgb24ToUV and bgr24ToUV 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 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" Archived-At: List-Archive: List-Post: T-Head C908: rgb24_to_uv_8_c: 2.7 rgb24_to_uv_8_rvv_i32: 3.2 rgb24_to_uv_128_c: 41.0 rgb24_to_uv_128_rvv_i32: 12.7 rgb24_to_uv_1080_c: 342.5 rgb24_to_uv_1080_rvv_i32: 105.7 rgb24_to_uv_1280_c: 406.0 rgb24_to_uv_1280_rvv_i32: 124.2 rgb24_to_uv_1920_c: 626.0 rgb24_to_uv_1920_rvv_i32: 186.0 SpacemiT X60: rgb24_to_uv_8_c: 2.5 rgb24_to_uv_8_rvv_i32: 3.0 rgb24_to_uv_128_c: 36.5 rgb24_to_uv_128_rvv_i32: 5.7 rgb24_to_uv_1080_c: 304.2 rgb24_to_uv_1080_rvv_i32: 49.0 rgb24_to_uv_1280_c: 360.5 rgb24_to_uv_1280_rvv_i32: 57.5 rgb24_to_uv_1920_c: 540.7 rgb24_to_uv_1920_rvv_i32: 86.2 --- libswscale/riscv/input.S | 46 ++++++++++++++++++++++++++++++++++++++ libswscale/riscv/swscale.c | 8 +++++++ 2 files changed, 54 insertions(+) diff --git a/libswscale/riscv/input.S b/libswscale/riscv/input.S index 323f650bc9..3392f189ca 100644 --- a/libswscale/riscv/input.S +++ b/libswscale/riscv/input.S @@ -53,3 +53,49 @@ func ff_rgb24ToY_rvv, zve32x ret endfunc + +func ff_bgr24ToUV_rvv, zve32x + lw t1, 20(a6) # BU + lw t4, 32(a6) # BV + lw t3, 12(a6) # RU + lw t6, 24(a6) # RV + j 1f +endfunc + +func ff_rgb24ToUV_rvv, zve32x + lw t1, 12(a6) # RU + lw t4, 24(a6) # RV + lw t3, 20(a6) # BU + lw t6, 32(a6) # BV +1: + lw t2, 16(a6) # GU + lw t5, 28(a6) # GV + li a7, (256 << (15 - 1)) + (1 << (15 - 7)) +2: + vsetvli t0, a5, e32, m8, ta, ma + vlseg3e8.v v0, (a3) + sub a5, a5, t0 + vzext.vf4 v16, v0 + sh1add a6, t0, t0 + vzext.vf4 v24, v2 + vmul.vx v8, v16, t1 + add a3, a6, a3 + vmul.vx v16, v16, t4 + vmacc.vx v8, t2, v24 + vmacc.vx v16, t5, v24 + vzext.vf4 v24, v4 + vadd.vx v8, v8, a7 + vadd.vx v16, v16, a7 + vmacc.vx v8, t3, v24 + vmacc.vx v16, t6, v24 + vsetvli zero, zero, e16, m4, ta, ma + vnsra.wi v0, v8, 15 - 6 + vnsra.wi v4, v16, 15 - 6 + vse16.v v0, (a0) + sh1add a0, t0, a0 + vse16.v v4, (a1) + sh1add a1, t0, a1 + bnez a5, 2b + + ret +endfunc diff --git a/libswscale/riscv/swscale.c b/libswscale/riscv/swscale.c index 187b3fce58..b3552976c6 100644 --- a/libswscale/riscv/swscale.c +++ b/libswscale/riscv/swscale.c @@ -23,8 +23,12 @@ 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_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 *); av_cold void ff_sws_init_swscale_riscv(SwsContext *c) { @@ -35,10 +39,14 @@ av_cold void ff_sws_init_swscale_riscv(SwsContext *c) switch (c->srcFormat) { case AV_PIX_FMT_BGR24: c->lumToYV12 = ff_bgr24ToY_rvv; + if (!c->chrSrcHSubSample) + c->chrToYV12 = ff_bgr24ToUV_rvv; break; case AV_PIX_FMT_RGB24: c->lumToYV12 = ff_rgb24ToY_rvv; + if (!c->chrSrcHSubSample) + c->chrToYV12 = ff_rgb24ToUV_rvv; break; } } -- 2.45.1 _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".