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 6693646360 for ; Thu, 15 Jun 2023 10:37:35 +0000 (UTC) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 31A0168C4D4; Thu, 15 Jun 2023 13:37:02 +0300 (EEST) Received: from azure-sdnproxy.icoremail.net (azure-sdnproxy.icoremail.net [52.237.72.81]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id A04FB68C462 for ; Thu, 15 Jun 2023 13:36:54 +0300 (EEST) Received: from host042-ubuntu-1804.lxd (unknown [10.12.130.38]) by app1 (Coremail) with SMTP id EwgMCgDnhMS_6Ypk6XcmAA--.44896S8; Thu, 15 Jun 2023 18:36:52 +0800 (CST) From: Peiting Shen To: ffmpeg-devel@ffmpeg.org Date: Thu, 15 Jun 2023 10:36:43 +0000 Message-Id: <20230615103645.25778-5-shenpeiting@eswincomputing.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20230615103645.25778-1-shenpeiting@eswincomputing.com> References: <20230615103645.25778-1-shenpeiting@eswincomputing.com> X-CM-TRANSID: EwgMCgDnhMS_6Ypk6XcmAA--.44896S8 X-Coremail-Antispam: 1UD129KBjvJXoWxWF47tFyDXFy5uFW3ArWUJwb_yoW5CrWxpr W7K343AF43Jr9xC3ZxGFyrZFWYqr1FgFn5JwnIv3W2vayqq34UGws2yryYyF15ZFyDAF13 WayS9a43Cr1UArJanT9S1TB71UUUUU7qnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnRJUUUBS14x267AKxVW5JVWrJwAFc2x0x2IEx4CE42xK8VAvwI8IcIk0 rVWrJVCq3wAFIxvE14AKwVWUJVWUGwA2048vs2IY020E87I2jVAFwI0_JF0E3s1l82xGYI kIc2x26xkF7I0E14v26r1I6r4UM28lY4IEw2IIxxk0rwA2F7IY1VAKz4vEj48ve4kI8wA2 z4x0Y4vE2Ix0cI8IcVAFwI0_tr0E3s1l84ACjcxK6xIIjxv20xvEc7CjxVAFwI0_Gr1j6F 4UJwA2z4x0Y4vEx4A2jsIE14v26rxl6s0DM28EF7xvwVC2z280aVCY1x0267AKxVW0oVCq 3wAS0I0E0xvYzxvE52x082IY62kv0487Mc02F40EFcxC0VAKzVAqx4xG6I80ewAv7VC0I7 IYx2IY67AKxVWUGVWUXwAv7VC2z280aVAFwI0_Gr0_Cr1lOx8S6xCaFVCjc4AY6r1j6r4U M4x0Y48IcxkI7VAKI48JM4x0x7Aq67IIx4CEVc8vx2IErcIFxwCY02Avz4vE-syl42xK82 IYc2Ij64vIr41l4I8I3I0E4IkC6x0Yz7v_Jr0_Gr1lx2IqxVAqx4xG67AKxVWUJVWUGwC2 0s026x8GjcxK67AKxVWUGVWUWwC2zVAF1VAY17CE14v26r1Y6r17MIIYrxkI7VAKI48JMI IF0xvE2Ix0cI8IcVAFwI0_Gr0_Xr1lIxAIcVC0I7IYx2IY6xkF7I0E14v26r4j6F4UMIIF 0xvE42xK8VAvwI8IcIk0rVWUJVWUCwCI42IY6I8E87Iv67AKxVW8JVWxJwCI42IY6I8E87 Iv6xkF7I0E14v26r4j6r4UJbIYCTnIWIevJa73UjIFyTuYvjfUedb1UUUUU X-CM-SenderInfo: hvkh01phlwx03j6h245lqf0zpsxwx03jof0z/ Subject: [FFmpeg-devel] [PATCH 4/6] lavc/ac3dsp: RISC-V V ac3_sum_square_butterfly_float 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: Shen Peiting MIME-Version: 1.0 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: From: Shen Peiting Scalar calculating float sum_square optimized by using RVV instructions Benchmarks on Spike(cycles): len=128 ac3_sum_square_butterfly_float_c: 7986 ac3_sum_square_butterfly_float_rvv: 146 len=1280 ac3_sum_square_butterfly_float_c: 79410 ac3_sum_square_butterfly_float_rvv: 1154 Co-Authored by: Yang Xiaojun Co-Authored by: Huang Xing Co-Authored by: Zeng Fanchen Signed-off-by: Shen Peiting --- libavcodec/riscv/ac3dsp_init.c | 6 ++++ libavcodec/riscv/ac3dsp_rvv.S | 54 ++++++++++++++++++++++++++++++++++ 2 files changed, 60 insertions(+) diff --git a/libavcodec/riscv/ac3dsp_init.c b/libavcodec/riscv/ac3dsp_init.c index 4fd4abe83e..d3aa20623a 100644 --- a/libavcodec/riscv/ac3dsp_init.c +++ b/libavcodec/riscv/ac3dsp_init.c @@ -30,6 +30,10 @@ void ff_ac3_sum_square_butterfly_int32_rvv(int64_t sum[4], const int32_t *coef0, const int32_t *coef1, int len); +void ff_ac3_sum_square_butterfly_float_rvv(float sum[4], + const float *coef0, + const float *coef1, + int len); av_cold void ff_ac3dsp_init_riscv(AC3DSPContext *c) { @@ -39,6 +43,8 @@ av_cold void ff_ac3dsp_init_riscv(AC3DSPContext *c) c->ac3_exponent_min = ff_ac3_exponent_min_rvv; c->float_to_fixed24 = ff_float_to_fixed24_rvv; } + if (flags & AV_CPU_FLAG_RVV_F32) + c->sum_square_butterfly_float = ff_ac3_sum_square_butterfly_float_rvv; #if (__riscv_xlen >= 64) if (flags & AV_CPU_FLAG_RVV_I64) c->sum_square_butterfly_int32 = ff_ac3_sum_square_butterfly_int32_rvv; diff --git a/libavcodec/riscv/ac3dsp_rvv.S b/libavcodec/riscv/ac3dsp_rvv.S index 4e0d238f85..05a4d44938 100644 --- a/libavcodec/riscv/ac3dsp_rvv.S +++ b/libavcodec/riscv/ac3dsp_rvv.S @@ -116,3 +116,57 @@ func ff_ac3_sum_square_butterfly_int32_rvv, zve64x addi a0, a0, 8 ret endfunc + + +func ff_ac3_sum_square_butterfly_float_rvv, zve32f + #Round Up + li t1, 0x61 + fscsr t1 + vsetvli t0, a3, e32, m4 + vle32.v v0, (a1) + vle32.v v4, (a2) + vfadd.vv v8, v0, v4 + vfsub.vv v12, v0, v4 + vfmul.vv v16, v0, v0 + vfmul.vv v20, v4, v4 + vfmul.vv v24, v8, v8 + vfmul.vv v28, v12, v12 + sub a3, a3, t0 + slli t0, t0, 2 + add a1, a1, t0 + add a2, a2, t0 + beq a3, x0, 2f +1: + vsetvli t0, a3, e32, m4 + vle32.v v0, (a1) + vle32.v v4, (a2) + vfadd.vv v8, v0, v4 + vfsub.vv v12, v0, v4 + vfmacc.vv v16, v0, v0 + vfmacc.vv v20, v4, v4 + vfmacc.vv v24, v8, v8 + vfmacc.vv v28, v12, v12 + sub a3, a3, t0 + slli t0, t0, 2 + add a1, a1, t0 + add a2, a2, t0 + bnez a3, 1b +2: + vsetvli t0, x0, e32, m4 + fcvt.s.w f0, x0 + vfmv.v.f v0, f0 + vfredsum.vs v0, v16, v0 + vfredsum.vs v1, v20, v1 + vfredsum.vs v2, v24, v2 + vfredsum.vs v3, v28, v3 + vsetivli t0, 1, e32, m1 + vse32.v v0, (a0) + addi a0, a0, 4 + vse32.v v1, (a0) + addi a0, a0, 4 + vse32.v v2, (a0) + addi a0, a0, 4 + vse32.v v3, (a0) + addi a0, a0, 4 + ret +endfunc -- 2.17.1 _______________________________________________ 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".