From patchwork Fri Sep 22 11:38:09 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Manojkumar Bhosale X-Patchwork-Id: 5236 Delivered-To: ffmpegpatchwork@gmail.com Received: by 10.2.36.26 with SMTP id f26csp3082322jaa; Fri, 22 Sep 2017 04:38:18 -0700 (PDT) X-Google-Smtp-Source: AOwi7QDZJj56JsmtVzwsFFtbmR8Fzgxlgnhu60hLKb8QSj+sl/qzt24/YZIrensiQgs4Us6EFOKI X-Received: by 10.223.172.239 with SMTP id o102mr4554383wrc.135.1506080298310; Fri, 22 Sep 2017 04:38:18 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1506080298; cv=none; d=google.com; s=arc-20160816; b=0/jGSfsEjubmOPO5hOjPWk3cm3pBj2mHcWZiGKPmd/8njRrhgWnwjbqCU+98RnAHhP XIovAi7LiFiBKpupPN5Ug9r/DdTFpr8lvTdH9vaPa+YXT9npoXc2dbg+DzszsYddXHdd 9wZvW6lawQ88pdkODF7Di8HbIIwhO7n57FOZQMMn4l3DRMHeBEzMr9IM8YbD79oK3snG 9HH1cQjONMM7SVLQ6MDrRQVYNCfKO6/n+UjNRjgiRV6W6mComNc0qJa9npqEjDuqCZFc GWQCrpCcx5QM0zb2TCGU94JPys2swaorsx7SQSeOn2+tcDBYk0xlIqh6QIzCX5ST6eZq 8RKw== 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:content-language :accept-language:in-reply-to:references:message-id:date:thread-index :thread-topic:to:from:delivered-to:arc-authentication-results; bh=RryqPBUIjYOtEU7xkIpMHfUqbJAzaLdcIinSkMsShhc=; b=a7bsLqCXprKAESOqPxlknTvvZihGW9NnoLvS6accXIRfD6/3xbtUkYt6FwqtqF7Zzo RCpQVthjFVfwIERwrsgCDJFT4syyN03jyJWs0zSZq6Pr4R4yLrgLfCqoiJktbVrr+sJZ XTP6M1dq+v9JO3LOV5EwwENf0WdSUAxkssSjvBP36HkUSoKrD2BurDPV84eSQr/pdm2s YhdP8cbXaUnTCl73mIYj82K1e8XgXU1RI6lzziu0OGDv0DDYJFZpXJtnC+OscMR5hPvN tLD+CEHsPlVwTO0MflyUnoc34+opZg5U8wwaMA2AEkyljYoxev76GOvUSAqFhPLNAEix vqpg== 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 b16si2888197wmd.92.2017.09.22.04.38.17; Fri, 22 Sep 2017 04:38:18 -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 CB312688364; Fri, 22 Sep 2017 14:38:06 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mailapp01.imgtec.com (mailapp01.imgtec.com [195.59.15.196]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id C2D61680790 for ; Fri, 22 Sep 2017 14:38:05 +0300 (EEST) Received: from hhmail02.hh.imgtec.org (unknown [10.100.10.20]) by Forcepoint Email with ESMTPS id 4ADFD4EFAF5EB for ; Fri, 22 Sep 2017 12:38:11 +0100 (IST) Received: from PUMAIL01.pu.imgtec.org (192.168.91.250) by hhmail02.hh.imgtec.org (10.100.10.20) with Microsoft SMTP Server (TLS) id 14.3.361.1; Fri, 22 Sep 2017 12:38:14 +0100 Received: from PUMAIL01.pu.imgtec.org ([::1]) by PUMAIL01.pu.imgtec.org ([::1]) with mapi id 14.03.0266.001; Fri, 22 Sep 2017 17:08:10 +0530 From: Manojkumar Bhosale To: FFmpeg development discussions and patches Thread-Topic: [FFmpeg-devel] [PATCH] avcodec/mips: Improve avc chroma horiz mc msa functions Thread-Index: AQHTMuHAxRV2xDrnFUiMF2o25zMRDqLAyLWg Date: Fri, 22 Sep 2017 11:38:09 +0000 Message-ID: <70293ACCC3BA6A4E81FFCA024C7A86E1E0592479@PUMAIL01.pu.imgtec.org> References: <1506002379-19302-1-git-send-email-kaustubh.raste@imgtec.com> In-Reply-To: <1506002379-19302-1-git-send-email-kaustubh.raste@imgtec.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [192.168.91.86] MIME-Version: 1.0 Subject: Re: [FFmpeg-devel] [PATCH] avcodec/mips: Improve avc chroma horiz mc msa functions X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.20 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: Kaustubh Raste Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" LGTM -----Original Message----- From: ffmpeg-devel [mailto:ffmpeg-devel-bounces@ffmpeg.org] On Behalf Of kaustubh.raste@imgtec.com Sent: Thursday, September 21, 2017 7:30 PM To: ffmpeg-devel@ffmpeg.org Cc: Kaustubh Raste Subject: [FFmpeg-devel] [PATCH] avcodec/mips: Improve avc chroma horiz mc msa functions From: Kaustubh Raste Replace generic with block size specific function. Signed-off-by: Kaustubh Raste --- libavcodec/mips/h264chroma_msa.c | 231 ++++++++++++++++++++++---------------- 1 file changed, 133 insertions(+), 98 deletions(-) mask = LD_SB(&chroma_mask_arr[0]); - LD_SB2(src, src_stride, src0, src1); + LD_SB2(src, stride, src0, src1); src0 = __msa_vshf_b(mask, src1, src0); res_r = __msa_dotp_u_h((v16u8) src0, coeff_vec); @@ -57,12 +56,11 @@ static void avc_chroma_hz_2x2_msa(uint8_t *src, int32_t src_stride, out1 = __msa_copy_u_h(res, 2); SH(out0, dst); - dst += dst_stride; + dst += stride; SH(out1, dst); } -static void avc_chroma_hz_2x4_msa(uint8_t *src, int32_t src_stride, - uint8_t *dst, int32_t dst_stride, +static void avc_chroma_hz_2x4_msa(uint8_t *src, uint8_t *dst, int32_t +stride, uint32_t coeff0, uint32_t coeff1) { v16u8 src0, src1, src2, src3; @@ -75,7 +73,7 @@ static void avc_chroma_hz_2x4_msa(uint8_t *src, int32_t src_stride, mask = LD_SB(&chroma_mask_arr[64]); - LD_UB4(src, src_stride, src0, src1, src2, src3); + LD_UB4(src, stride, src0, src1, src2, src3); VSHF_B2_UB(src0, src1, src2, src3, mask, mask, src0, src2); @@ -87,64 +85,21 @@ static void avc_chroma_hz_2x4_msa(uint8_t *src, int32_t src_stride, res_r = __msa_sat_u_h(res_r, 7); res = (v8i16) __msa_pckev_b((v16i8) res_r, (v16i8) res_r); - ST2x4_UB(res, 0, dst, dst_stride); -} - -static void avc_chroma_hz_2x8_msa(uint8_t *src, int32_t src_stride, - uint8_t *dst, int32_t dst_stride, - uint32_t coeff0, uint32_t coeff1) -{ - v16u8 src0, src1, src2, src3, src4, src5, src6, src7; - v8u16 res_r; - v8i16 res; - v16i8 mask; - v16i8 coeff_vec0 = __msa_fill_b(coeff0); - v16i8 coeff_vec1 = __msa_fill_b(coeff1); - v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1); - - mask = LD_SB(&chroma_mask_arr[64]); - - LD_UB8(src, src_stride, src0, src1, src2, src3, src4, src5, src6, src7); - - VSHF_B2_UB(src0, src1, src2, src3, mask, mask, src0, src2); - VSHF_B2_UB(src4, src5, src6, src7, mask, mask, src4, src6); - - ILVR_D2_UB(src2, src0, src6, src4, src0, src4); - - res_r = __msa_dotp_u_h(src0, coeff_vec); - res_r <<= 3; - res_r = (v8u16) __msa_srari_h((v8i16) res_r, 6); - res_r = __msa_sat_u_h(res_r, 7); - res = (v8i16) __msa_pckev_b((v16i8) res_r, (v16i8) res_r); - - ST2x4_UB(res, 0, dst, dst_stride); - dst += (4 * dst_stride); - - res_r = __msa_dotp_u_h(src4, coeff_vec); - res_r <<= 3; - res_r = (v8u16) __msa_srari_h((v8i16) res_r, 6); - res_r = __msa_sat_u_h(res_r, 7); - res = (v8i16) __msa_pckev_b((v16i8) res_r, (v16i8) res_r); - - ST2x4_UB(res, 0, dst, dst_stride); + ST2x4_UB(res, 0, dst, stride); } -static void avc_chroma_hz_2w_msa(uint8_t *src, int32_t src_stride, - uint8_t *dst, int32_t dst_stride, +static void avc_chroma_hz_2w_msa(uint8_t *src, uint8_t *dst, int32_t +stride, uint32_t coeff0, uint32_t coeff1, int32_t height) { if (2 == height) { - avc_chroma_hz_2x2_msa(src, src_stride, dst, dst_stride, coeff0, coeff1); + avc_chroma_hz_2x2_msa(src, dst, stride, coeff0, coeff1); } else if (4 == height) { - avc_chroma_hz_2x4_msa(src, src_stride, dst, dst_stride, coeff0, coeff1); - } else if (8 == height) { - avc_chroma_hz_2x8_msa(src, src_stride, dst, dst_stride, coeff0, coeff1); + avc_chroma_hz_2x4_msa(src, dst, stride, coeff0, coeff1); } } -static void avc_chroma_hz_4x2_msa(uint8_t *src, int32_t src_stride, - uint8_t *dst, int32_t dst_stride, +static void avc_chroma_hz_4x2_msa(uint8_t *src, uint8_t *dst, int32_t +stride, uint32_t coeff0, uint32_t coeff1) { v16i8 src0, src1; @@ -157,7 +112,7 @@ static void avc_chroma_hz_4x2_msa(uint8_t *src, int32_t src_stride, mask = LD_SB(&chroma_mask_arr[0]); - LD_SB2(src, src_stride, src0, src1); + LD_SB2(src, stride, src0, src1); src0 = __msa_vshf_b(mask, src1, src0); res_r = __msa_dotp_u_h((v16u8) src0, coeff_vec); @@ -166,18 +121,14 @@ static void avc_chroma_hz_4x2_msa(uint8_t *src, int32_t src_stride, res_r = __msa_sat_u_h(res_r, 7); res = (v4i32) __msa_pckev_b((v16i8) res_r, (v16i8) res_r); - ST4x2_UB(res, dst, dst_stride); + ST4x2_UB(res, dst, stride); } -static void avc_chroma_hz_4x4multiple_msa(uint8_t *src, int32_t src_stride, - uint8_t *dst, int32_t dst_stride, - uint32_t coeff0, uint32_t coeff1, - int32_t height) +static void avc_chroma_hz_4x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, + uint32_t coeff0, uint32_t coeff1) { - uint32_t row; - v16u8 src0, src1, src2, src3; + v16u8 src0, src1, src2, src3, out; v8u16 res0_r, res1_r; - v4i32 res0, res1; v16i8 mask; v16i8 coeff_vec0 = __msa_fill_b(coeff0); v16i8 coeff_vec1 = __msa_fill_b(coeff1); @@ -185,42 +136,113 @@ static void avc_chroma_hz_4x4multiple_msa(uint8_t *src, int32_t src_stride, mask = LD_SB(&chroma_mask_arr[0]); - for (row = (height >> 2); row--;) { - LD_UB4(src, src_stride, src0, src1, src2, src3); - src += (4 * src_stride); - - VSHF_B2_UB(src0, src1, src2, src3, mask, mask, src0, src2); - DOTP_UB2_UH(src0, src2, coeff_vec, coeff_vec, res0_r, res1_r); + LD_UB4(src, stride, src0, src1, src2, src3); + VSHF_B2_UB(src0, src1, src2, src3, mask, mask, src0, src2); + DOTP_UB2_UH(src0, src2, coeff_vec, coeff_vec, res0_r, res1_r); + res0_r <<= 3; + res1_r <<= 3; + SRARI_H2_UH(res0_r, res1_r, 6); + SAT_UH2_UH(res0_r, res1_r, 7); + out = (v16u8) __msa_pckev_b((v16i8) res1_r, (v16i8) res0_r); + ST4x4_UB(out, out, 0, 1, 2, 3, dst, stride); } - res0_r <<= 3; - res1_r <<= 3; +static void avc_chroma_hz_4x8_msa(uint8_t *src, uint8_t *dst, int32_t stride, + uint32_t coeff0, uint32_t coeff1) { + v16u8 src0, src1, src2, src3, src4, src5, src6, src7, out0, out1; + v16i8 mask; + v8u16 res0, res1, res2, res3; + v16i8 coeff_vec0 = __msa_fill_b(coeff0); + v16i8 coeff_vec1 = __msa_fill_b(coeff1); + v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1); - SRARI_H2_UH(res0_r, res1_r, 6); - SAT_UH2_UH(res0_r, res1_r, 7); - PCKEV_B2_SW(res0_r, res0_r, res1_r, res1_r, res0, res1); + mask = LD_SB(&chroma_mask_arr[0]); - ST4x4_UB(res0, res1, 0, 1, 0, 1, dst, dst_stride); - dst += (4 * dst_stride); - } + LD_UB8(src, stride, src0, src1, src2, src3, src4, src5, src6, src7); + VSHF_B2_UB(src0, src1, src2, src3, mask, mask, src0, src2); + VSHF_B2_UB(src4, src5, src6, src7, mask, mask, src4, src6); + DOTP_UB2_UH(src0, src2, coeff_vec, coeff_vec, res0, res1); + DOTP_UB2_UH(src4, src6, coeff_vec, coeff_vec, res2, res3); + SLLI_4V(res0, res1, res2, res3, 3); + SRARI_H4_UH(res0, res1, res2, res3, 6); + SAT_UH4_UH(res0, res1, res2, res3, 7); + PCKEV_B2_UB(res1, res0, res3, res2, out0, out1); + ST4x8_UB(out0, out1, dst, stride); } -static void avc_chroma_hz_4w_msa(uint8_t *src, int32_t src_stride, - uint8_t *dst, int32_t dst_stride, +static void avc_chroma_hz_4w_msa(uint8_t *src, uint8_t *dst, int32_t +stride, uint32_t coeff0, uint32_t coeff1, int32_t height) { if (2 == height) { - avc_chroma_hz_4x2_msa(src, src_stride, dst, dst_stride, coeff0, coeff1); - } else { - avc_chroma_hz_4x4multiple_msa(src, src_stride, dst, dst_stride, coeff0, - coeff1, height); + avc_chroma_hz_4x2_msa(src, dst, stride, coeff0, coeff1); + } else if (4 == height) { + avc_chroma_hz_4x4_msa(src, dst, stride, coeff0, coeff1); + } else if (8 == height) { + avc_chroma_hz_4x8_msa(src, dst, stride, coeff0, coeff1); } } -static void avc_chroma_hz_8w_msa(uint8_t *src, int32_t src_stride, - uint8_t *dst, int32_t dst_stride, - uint32_t coeff0, uint32_t coeff1, - int32_t height) +static void avc_chroma_hz_8x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, + uint32_t coeff0, uint32_t coeff1) { + v16u8 src0, src1, src2, src3, out0, out1; + v8u16 res0, res1, res2, res3; + v16i8 mask; + v16i8 coeff_vec0 = __msa_fill_b(coeff0); + v16i8 coeff_vec1 = __msa_fill_b(coeff1); + v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1); + + mask = LD_SB(&chroma_mask_arr[32]); + LD_UB4(src, stride, src0, src1, src2, src3); + VSHF_B2_UB(src0, src0, src1, src1, mask, mask, src0, src1); + VSHF_B2_UB(src2, src2, src3, src3, mask, mask, src2, src3); + DOTP_UB4_UH(src0, src1, src2, src3, coeff_vec, coeff_vec, coeff_vec, + coeff_vec, res0, res1, res2, res3); + SLLI_4V(res0, res1, res2, res3, 3); + SRARI_H4_UH(res0, res1, res2, res3, 6); + SAT_UH4_UH(res0, res1, res2, res3, 7); + PCKEV_B2_UB(res1, res0, res3, res2, out0, out1); + ST8x4_UB(out0, out1, dst, stride); +} + +static void avc_chroma_hz_8x8_msa(uint8_t *src, uint8_t *dst, int32_t stride, + uint32_t coeff0, uint32_t coeff1) { + v16u8 src0, src1, src2, src3, src4, src5, src6, src7; + v16u8 out0, out1, out2, out3; + v8u16 res0, res1, res2, res3, res4, res5, res6, res7; + v16i8 mask; + v16i8 coeff_vec0 = __msa_fill_b(coeff0); + v16i8 coeff_vec1 = __msa_fill_b(coeff1); + v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1); + + mask = LD_SB(&chroma_mask_arr[32]); + + LD_UB8(src, stride, src0, src1, src2, src3, src4, src5, src6, src7); + VSHF_B2_UB(src0, src0, src1, src1, mask, mask, src0, src1); + VSHF_B2_UB(src2, src2, src3, src3, mask, mask, src2, src3); + VSHF_B2_UB(src4, src4, src5, src5, mask, mask, src4, src5); + VSHF_B2_UB(src6, src6, src7, src7, mask, mask, src6, src7); + DOTP_UB4_UH(src0, src1, src2, src3, coeff_vec, coeff_vec, coeff_vec, + coeff_vec, res0, res1, res2, res3); + DOTP_UB4_UH(src4, src5, src6, src7, coeff_vec, coeff_vec, coeff_vec, + coeff_vec, res4, res5, res6, res7); + SLLI_4V(res0, res1, res2, res3, 3); + SLLI_4V(res4, res5, res6, res7, 3); + SRARI_H4_UH(res0, res1, res2, res3, 6); + SRARI_H4_UH(res4, res5, res6, res7, 6); + SAT_UH4_UH(res0, res1, res2, res3, 7); + SAT_UH4_UH(res4, res5, res6, res7, 7); + PCKEV_B2_UB(res1, res0, res3, res2, out0, out1); + PCKEV_B2_UB(res5, res4, res7, res6, out2, out3); + ST8x8_UB(out0, out1, out2, out3, dst, stride); } + +static void avc_chroma_hz_nonmult_msa(uint8_t *src, uint8_t *dst, + int32_t stride, uint32_t coeff0, + uint32_t coeff1, int32_t height) { uint32_t row; v16u8 src0, src1, src2, src3, out0, out1; @@ -233,8 +255,8 @@ static void avc_chroma_hz_8w_msa(uint8_t *src, int32_t src_stride, mask = LD_SB(&chroma_mask_arr[32]); for (row = height >> 2; row--;) { - LD_UB4(src, src_stride, src0, src1, src2, src3); - src += (4 * src_stride); + LD_UB4(src, stride, src0, src1, src2, src3); + src += (4 * stride); VSHF_B2_UB(src0, src0, src1, src1, mask, mask, src0, src1); VSHF_B2_UB(src2, src2, src3, src3, mask, mask, src2, src3); @@ -244,14 +266,14 @@ static void avc_chroma_hz_8w_msa(uint8_t *src, int32_t src_stride, SRARI_H4_UH(res0, res1, res2, res3, 6); SAT_UH4_UH(res0, res1, res2, res3, 7); PCKEV_B2_UB(res1, res0, res3, res2, out0, out1); - ST8x4_UB(out0, out1, dst, dst_stride); - dst += (4 * dst_stride); + ST8x4_UB(out0, out1, dst, stride); + dst += (4 * stride); } if (0 != (height % 4)) { for (row = (height % 4); row--;) { src0 = LD_UB(src); - src += src_stride; + src += stride; src0 = (v16u8) __msa_vshf_b(mask, (v16i8) src0, (v16i8) src0); @@ -262,11 +284,24 @@ static void avc_chroma_hz_8w_msa(uint8_t *src, int32_t src_stride, res0 = (v8u16) __msa_pckev_b((v16i8) res0, (v16i8) res0); ST8x1_UB(res0, dst); - dst += dst_stride; + dst += stride; } } } +static void avc_chroma_hz_8w_msa(uint8_t *src, uint8_t *dst, int32_t stride, + uint32_t coeff0, uint32_t coeff1, + int32_t height) { + if (4 == height) { + avc_chroma_hz_8x4_msa(src, dst, stride, coeff0, coeff1); + } else if (8 == height) { + avc_chroma_hz_8x8_msa(src, dst, stride, coeff0, coeff1); + } else { + avc_chroma_hz_nonmult_msa(src, dst, stride, coeff0, coeff1, height); + } +} + static void avc_chroma_vt_2x2_msa(uint8_t *src, int32_t src_stride, uint8_t *dst, int32_t dst_stride, uint32_t coeff0, uint32_t coeff1) @@ -1877,7 +1912,7 @@ void ff_put_h264_chroma_mc8_msa(uint8_t *dst, uint8_t *src, avc_chroma_hv_8w_msa(src, stride, dst, stride, x, (8 - x), y, (8 - y), height); } else if (x) { - avc_chroma_hz_8w_msa(src, stride, dst, stride, x, (8 - x), height); + avc_chroma_hz_8w_msa(src, dst, stride, x, (8 - x), height); } else if (y) { avc_chroma_vt_8w_msa(src, stride, dst, stride, y, (8 - y), height); } else { @@ -1896,7 +1931,7 @@ void ff_put_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src, avc_chroma_hv_4w_msa(src, stride, dst, stride, x, (8 - x), y, (8 - y), height); } else if (x) { - avc_chroma_hz_4w_msa(src, stride, dst, stride, x, (8 - x), height); + avc_chroma_hz_4w_msa(src, dst, stride, x, (8 - x), height); } else if (y) { avc_chroma_vt_4w_msa(src, stride, dst, stride, y, (8 - y), height); } else { @@ -1920,7 +1955,7 @@ void ff_put_h264_chroma_mc2_msa(uint8_t *dst, uint8_t *src, avc_chroma_hv_2w_msa(src, stride, dst, stride, x, (8 - x), y, (8 - y), height); } else if (x) { - avc_chroma_hz_2w_msa(src, stride, dst, stride, x, (8 - x), height); + avc_chroma_hz_2w_msa(src, dst, stride, x, (8 - x), height); } else if (y) { avc_chroma_vt_2w_msa(src, stride, dst, stride, y, (8 - y), height); } else { -- 1.7.9.5 _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel diff --git a/libavcodec/mips/h264chroma_msa.c b/libavcodec/mips/h264chroma_msa.c index 940e12d..c27830d 100644 --- a/libavcodec/mips/h264chroma_msa.c +++ b/libavcodec/mips/h264chroma_msa.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2015 Shivraj Patil (Shivraj.Patil@imgtec.com) + * Copyright (c) 2015 - 2017 Shivraj Patil (Shivraj.Patil@imgtec.com) * * This file is part of FFmpeg. * @@ -29,8 +29,7 @@ static const uint8_t chroma_mask_arr[16 * 5] = { 0, 1, 1, 2, 16, 17, 17, 18, 16, 17, 17, 18, 18, 19, 19, 20 }; -static void avc_chroma_hz_2x2_msa(uint8_t *src, int32_t src_stride, - uint8_t *dst, int32_t dst_stride, +static void avc_chroma_hz_2x2_msa(uint8_t *src, uint8_t *dst, int32_t +stride, uint32_t coeff0, uint32_t coeff1) { uint16_t out0, out1; @@ -44,7 +43,7 @@ static void avc_chroma_hz_2x2_msa(uint8_t *src, int32_t src_stride,