From patchwork Wed Sep 5 10:31:05 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Shiyou Yin X-Patchwork-Id: 10238 Delivered-To: ffmpegpatchwork@gmail.com Received: by 2002:a02:12c4:0:0:0:0:0 with SMTP id 65-v6csp4948221jap; Wed, 5 Sep 2018 03:31:22 -0700 (PDT) X-Google-Smtp-Source: ANB0VdacDtg2V7aXqB9Zbxpx4QCbx4IOCiQsanB9hskRcJuzQV3Hmh+TkZGrFrxGP0ykOCfRaXvf X-Received: by 2002:a1c:c7c3:: with SMTP id x186-v6mr10421780wmf.109.1536143482712; Wed, 05 Sep 2018 03:31:22 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1536143482; cv=none; d=google.com; s=arc-20160816; b=RC8LW/AQm+pHwYUCUzT9yIc3gSGX991819ENpG+qfg6dIrxUweNBJFwZ3//VgO3JFQ 4Hj/OU7bLZvVhM6mOSuzJJeoipdn85+KgKBQSTCONDXzwILxeSLKF1S2flSMsuGfNZUn yRVjMxQ7KDqKLOCoYcN5CEEzKgdKLx6eHvUeezhooKQ6M+uFLPAvm3VAfy2YSLpkAA4a ylBPPY8NAKZrMoX/J4mBQEfBXMYBm2M1BcM42n98Ry1zkRxtTz7lz6tQSl/fV2HKBMGP XVollB3teOehwLSrKmAzB5/lCe8JIQvA4wTEQ8GR2suWvAImAioFHHOh2SvsvVgXv6NG oslw== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:mime-version:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:message-id:date:to:from:delivered-to; bh=BcCk1onDlc6hRgqwu6va9THXP9wezSKIIJZYICD5EOA=; b=KIR4nezOrlk/Nej1v3QAmjyxLathnnp4BfGIWnvAm8QYswkEGO2uZ8Bz9SKnNMat1r 36JcCzYrtl2+XqX0cZjAOvkUvMIZGaZCCALLUxycU+tOiF2uwSm3Kk9jVV2Kl1J3hrkN MOFS5xj8AdZa3sfwlSsf4uY8rPMvXpBbYk+JRH+NMdoqXnK+8tzOqHSDWdKhpkECccrG PWnTNyTDD6xBKX9+Vcf15uIPEBCQZma19Yn2YKUR3wYiabvhkUg4AQcoguxc38NVjR47 4o+3Cyo16e1exehxgrt+c4D+R68l4qpskB2uVawKmzwVHwPSzVCRi4/CK1azOyJISJTl XQWg== 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 g14-v6si1421446wmh.93.2018.09.05.03.31.21; Wed, 05 Sep 2018 03:31:22 -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 D6252689F50; Wed, 5 Sep 2018 13:31:12 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail.loongson.cn (mail.loongson.cn [114.242.206.163]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 12494680A24 for ; Wed, 5 Sep 2018 13:31:03 +0300 (EEST) Received: from localhost (unknown [210.45.123.188]) by mail (Coremail) with SMTP id QMiowPCxOeVNsI9be3MOAA--.360S3; Wed, 05 Sep 2018 18:30:38 +0800 (CST) From: Shiyou Yin To: ffmpeg-devel@ffmpeg.org Date: Wed, 5 Sep 2018 18:31:05 +0800 Message-Id: <1536143466-8610-1-git-send-email-yinshiyou-hf@loongson.cn> X-Mailer: git-send-email 2.1.0 X-CM-TRANSID: QMiowPCxOeVNsI9be3MOAA--.360S3 X-Coremail-Antispam: 1UD129KBjvJXoW3ZF48KrWrXF18CF45Gr4xXrb_yoWDWFWkp3 y8ArnYva12yFW5W3W7Jr1UCr9xZr4DtFsFkrWq9F92qwn5tFn7XrWxKr95GrWYkFyYvF4U Ar1DAa4rAwn5Gr7anT9S1TB71UUUUU7qnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnRJUUUkvb4IE77IF4wAFF20E14v26r1j6r4UM7CY07I20VC2zVCF04k2 6cxKx2IYs7xG6rWj6s0DM7CIcVAFz4kK6r1j6r18M28lY4IEw2IIxxk0rwA2F7IY1VAKz4 vEj48ve4kI8wA2z4x0Y4vE2Ix0cI8IcVAFwI0_JFI_Gr1l84ACjcxK6xIIjxv20xvEc7Cj xVAFwI0_Gr0_Cr1l84ACjcxK6I8E87Iv67AKxVW8Jr0_Cr1UM28EF7xvwVC2z280aVCY1x 0267AKxVWxJr0_GcWle2I262IYc4CY6c8Ij28IcVAaY2xG8wAqx4xG64xvF2IEw4CE5I8C rVC2j2WlYx0E2Ix0cI8IcVAFwI0_Jrv_JF1lYx0Ex4A2jsIE14v26r4j6F4UMcvjeVCFs4 IE7xkEbVWUJVW8JwACjcxG0xvY0x0EwIxGrwACjcxG0xvY0x0EwIxGrVCF72vEw4AK0wCY 02Avz4vE14v_Gw1l42xK82IYc2Ij64vIr41l4I8I3I0E4IkC6x0Yz7v_Jr0_Gr1lx2IqxV Aqx4xG67AKxVWUJVWUGwC20s026x8GjcxK67AKxVWUGVWUWwC2zVAF1VAY17CE14v26r1j 6r15MIIF0xvE2Ix0cI8IcVAFwI0_Jr0_JF4lIxAIcVC0I7IYx2IY6xkF7I0E14v26r1j6r 4UMIIF0xvE42xK8VAvwI8IcIk0rVWrJr0_WFyUJwCI42IY6I8E87Iv67AKxVWUJVW8JwCI 42IY6I8E87Iv6xkF7I0E14v26r1j6r4UYxBIdaVFxhVjvjDU0xZFpf9x07jOsqXUUUUU= X-CM-SenderInfo: p1lq2x5l1r3gtki6z05rqj20fqof0/ Subject: [FFmpeg-devel] [PATCH 1/2] avcodec/mips: [loongson] reoptimize put and add pixels clamped 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 MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" Simplify the usage of intermediate variable addr and remove unused variable all64 in following functions: 1. ff_put_pixels_clamped_mmi 2. ff_put_signed_pixels_clamped_mmi 3. ff_add_pixels_clamped_mmi This optimization speed up mpeg4 decode about 2% on loongson platform(tested with 3A3000). --- libavcodec/mips/idctdsp_mmi.c | 155 +++++++++++++++++------------------------- 1 file changed, 62 insertions(+), 93 deletions(-) diff --git a/libavcodec/mips/idctdsp_mmi.c b/libavcodec/mips/idctdsp_mmi.c index b797965..a96dac4 100644 --- a/libavcodec/mips/idctdsp_mmi.c +++ b/libavcodec/mips/idctdsp_mmi.c @@ -29,9 +29,6 @@ void ff_put_pixels_clamped_mmi(const int16_t *block, uint8_t *av_restrict pixels, ptrdiff_t line_size) { double ftmp[8]; - mips_reg addr[1]; - DECLARE_VAR_ALL64; - DECLARE_VAR_ADDRT; __asm__ volatile ( MMI_LDC1(%[ftmp0], %[block], 0x00) @@ -42,60 +39,44 @@ void ff_put_pixels_clamped_mmi(const int16_t *block, MMI_LDC1(%[ftmp5], %[block], 0x28) MMI_LDC1(%[ftmp6], %[block], 0x30) MMI_LDC1(%[ftmp7], %[block], 0x38) - PTR_ADDU "%[addr0], %[pixels], %[line_size] \n\t" "packushb %[ftmp0], %[ftmp0], %[ftmp1] \n\t" "packushb %[ftmp2], %[ftmp2], %[ftmp3] \n\t" "packushb %[ftmp4], %[ftmp4], %[ftmp5] \n\t" "packushb %[ftmp6], %[ftmp6], %[ftmp7] \n\t" MMI_SDC1(%[ftmp0], %[pixels], 0x00) - MMI_SDC1(%[ftmp2], %[addr0], 0x00) - MMI_SDXC1(%[ftmp4], %[addr0], %[line_size], 0x00) - MMI_SDXC1(%[ftmp6], %[pixels], %[line_sizex3], 0x00) - : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), - [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), - [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), - [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), - RESTRICT_ASM_ALL64 - RESTRICT_ASM_ADDRT - [addr0]"=&r"(addr[0]), - [pixels]"+&r"(pixels) - : [line_size]"r"((mips_reg)line_size), - [line_sizex3]"r"((mips_reg)(line_size*3)), - [block]"r"(block) - : "memory" - ); - - pixels += line_size*4; - block += 32; + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp2], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp4], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp6], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" - __asm__ volatile ( - MMI_LDC1(%[ftmp0], %[block], 0x00) - MMI_LDC1(%[ftmp1], %[block], 0x08) - MMI_LDC1(%[ftmp2], %[block], 0x10) - MMI_LDC1(%[ftmp3], %[block], 0x18) - MMI_LDC1(%[ftmp4], %[block], 0x20) - MMI_LDC1(%[ftmp5], %[block], 0x28) - MMI_LDC1(%[ftmp6], %[block], 0x30) - MMI_LDC1(%[ftmp7], %[block], 0x38) - PTR_ADDU "%[addr0], %[pixels], %[line_size] \n\t" + MMI_LDC1(%[ftmp0], %[block], 0x40) + MMI_LDC1(%[ftmp1], %[block], 0x48) + MMI_LDC1(%[ftmp2], %[block], 0x50) + MMI_LDC1(%[ftmp3], %[block], 0x58) + MMI_LDC1(%[ftmp4], %[block], 0x60) + MMI_LDC1(%[ftmp5], %[block], 0x68) + MMI_LDC1(%[ftmp6], %[block], 0x70) + MMI_LDC1(%[ftmp7], %[block], 0x78) "packushb %[ftmp0], %[ftmp0], %[ftmp1] \n\t" "packushb %[ftmp2], %[ftmp2], %[ftmp3] \n\t" "packushb %[ftmp4], %[ftmp4], %[ftmp5] \n\t" "packushb %[ftmp6], %[ftmp6], %[ftmp7] \n\t" MMI_SDC1(%[ftmp0], %[pixels], 0x00) - MMI_SDC1(%[ftmp2], %[addr0], 0x00) - MMI_SDXC1(%[ftmp4], %[addr0], %[line_size], 0x00) - MMI_SDXC1(%[ftmp6], %[pixels], %[line_sizex3], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp2], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp4], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp6], %[pixels], 0x00) : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), - RESTRICT_ASM_ALL64 - RESTRICT_ASM_ADDRT - [addr0]"=&r"(addr[0]), [pixels]"+&r"(pixels) : [line_size]"r"((mips_reg)line_size), - [line_sizex3]"r"((mips_reg)(line_size*3)), [block]"r"(block) : "memory" ); @@ -104,15 +85,9 @@ void ff_put_pixels_clamped_mmi(const int16_t *block, void ff_put_signed_pixels_clamped_mmi(const int16_t *block, uint8_t *av_restrict pixels, ptrdiff_t line_size) { - int64_t line_skip = line_size; - int64_t line_skip3 = 0; double ftmp[5]; - mips_reg addr[1]; - DECLARE_VAR_ALL64; - DECLARE_VAR_ADDRT; __asm__ volatile ( - PTR_ADDU "%[line_skip3], %[line_skip], %[line_skip] \n\t" MMI_LDC1(%[ftmp1], %[block], 0x00) MMI_LDC1(%[ftmp0], %[block], 0x08) "packsshb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" @@ -130,12 +105,14 @@ void ff_put_signed_pixels_clamped_mmi(const int16_t *block, "paddb %[ftmp3], %[ftmp3], %[ff_pb_80] \n\t" "paddb %[ftmp4], %[ftmp4], %[ff_pb_80] \n\t" MMI_SDC1(%[ftmp1], %[pixels], 0x00) - MMI_SDXC1(%[ftmp2], %[pixels], %[line_skip], 0x00) - MMI_SDXC1(%[ftmp3], %[pixels], %[line_skip3], 0x00) - PTR_ADDU "%[line_skip3], %[line_skip3], %[line_skip] \n\t" - MMI_SDXC1(%[ftmp4], %[pixels], %[line_skip3], 0x00) - PTR_ADDU "%[addr0], %[line_skip3], %[line_skip] \n\t" - PTR_ADDU "%[pixels], %[pixels], %[addr0] \n\t" + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp2], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp3], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp4], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_LDC1(%[ftmp1], %[block], 0x40) MMI_LDC1(%[ftmp0], %[block], 0x48) "packsshb %[ftmp1], %[ftmp1], %[ftmp0] \n\t" @@ -153,19 +130,18 @@ void ff_put_signed_pixels_clamped_mmi(const int16_t *block, "paddb %[ftmp3], %[ftmp3], %[ff_pb_80] \n\t" "paddb %[ftmp4], %[ftmp4], %[ff_pb_80] \n\t" MMI_SDC1(%[ftmp1], %[pixels], 0x00) - MMI_SDXC1(%[ftmp2], %[pixels], %[line_skip], 0x00) - PTR_ADDU "%[addr0], %[line_skip], %[line_skip] \n\t" - MMI_SDXC1(%[ftmp3], %[pixels], %[addr0], 0x00) - MMI_SDXC1(%[ftmp4], %[pixels], %[line_skip3], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp2], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp3], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp4], %[pixels], 0x00) : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), [ftmp4]"=&f"(ftmp[4]), - RESTRICT_ASM_ALL64 - RESTRICT_ASM_ADDRT - [addr0]"=&r"(addr[0]), - [pixels]"+&r"(pixels), [line_skip3]"+&r"(line_skip3) + [pixels]"+&r"(pixels) : [block]"r"(block), - [line_skip]"r"((mips_reg)line_skip), + [line_size]"r"((mips_reg)line_size), [ff_pb_80]"f"(ff_pb_80) : "memory" ); @@ -174,49 +150,42 @@ void ff_put_signed_pixels_clamped_mmi(const int16_t *block, void ff_add_pixels_clamped_mmi(const int16_t *block, uint8_t *av_restrict pixels, ptrdiff_t line_size) { - double ftmp[8]; + double ftmp[9]; uint64_t tmp[1]; - mips_reg addr[1]; - DECLARE_VAR_ALL64; - DECLARE_VAR_ADDRT; - __asm__ volatile ( - "li %[tmp0], 0x04 \n\t" - "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" - "1: \n\t" + "li %[tmp0], 0x04 \n\t" + "xor %[ftmp0], %[ftmp0], %[ftmp0] \n\t" + "1: \n\t" + MMI_LDC1(%[ftmp5], %[pixels], 0x00) + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_LDC1(%[ftmp6], %[pixels], 0x00) + PTR_SUBU "%[pixels], %[pixels], %[line_size] \n\t" MMI_LDC1(%[ftmp1], %[block], 0x00) MMI_LDC1(%[ftmp2], %[block], 0x08) MMI_LDC1(%[ftmp3], %[block], 0x10) MMI_LDC1(%[ftmp4], %[block], 0x18) - MMI_LDC1(%[ftmp5], %[pixels], 0x00) - MMI_LDXC1(%[ftmp6], %[pixels], %[line_size], 0x00) - "mov.d %[ftmp7], %[ftmp5] \n\t" - "punpcklbh %[ftmp5], %[ftmp5], %[ftmp0] \n\t" - "punpckhbh %[ftmp7], %[ftmp7], %[ftmp0] \n\t" - "paddh %[ftmp1], %[ftmp1], %[ftmp5] \n\t" - "paddh %[ftmp2], %[ftmp2], %[ftmp7] \n\t" - "mov.d %[ftmp7], %[ftmp6] \n\t" - "punpcklbh %[ftmp6], %[ftmp6], %[ftmp0] \n\t" - "punpckhbh %[ftmp7], %[ftmp7], %[ftmp0] \n\t" - "paddh %[ftmp3], %[ftmp3], %[ftmp6] \n\t" - "paddh %[ftmp4], %[ftmp4], %[ftmp7] \n\t" - "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" - "packushb %[ftmp3], %[ftmp3], %[ftmp4] \n\t" + PTR_ADDIU "%[block], %[block], 0x20 \n\t" + "punpckhbh %[ftmp7], %[ftmp5], %[ftmp0] \n\t" + "punpcklbh %[ftmp5], %[ftmp5], %[ftmp0] \n\t" + "punpckhbh %[ftmp8], %[ftmp6], %[ftmp0] \n\t" + "punpcklbh %[ftmp6], %[ftmp6], %[ftmp0] \n\t" + "paddh %[ftmp1], %[ftmp1], %[ftmp5] \n\t" + "paddh %[ftmp2], %[ftmp2], %[ftmp7] \n\t" + "paddh %[ftmp3], %[ftmp3], %[ftmp6] \n\t" + "paddh %[ftmp4], %[ftmp4], %[ftmp8] \n\t" + "packushb %[ftmp1], %[ftmp1], %[ftmp2] \n\t" + "packushb %[ftmp3], %[ftmp3], %[ftmp4] \n\t" MMI_SDC1(%[ftmp1], %[pixels], 0x00) - MMI_SDXC1(%[ftmp3], %[pixels], %[line_size], 0x00) - "addi %[tmp0], %[tmp0], -0x01 \n\t" - PTR_ADDIU "%[block], %[block], 0x20 \n\t" - PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" - PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" - "bnez %[tmp0], 1b" + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + MMI_SDC1(%[ftmp3], %[pixels], 0x00) + "addi %[tmp0], %[tmp0], -0x01 \n\t" + PTR_ADDU "%[pixels], %[pixels], %[line_size] \n\t" + "bnez %[tmp0], 1b \n\t" : [ftmp0]"=&f"(ftmp[0]), [ftmp1]"=&f"(ftmp[1]), [ftmp2]"=&f"(ftmp[2]), [ftmp3]"=&f"(ftmp[3]), [ftmp4]"=&f"(ftmp[4]), [ftmp5]"=&f"(ftmp[5]), [ftmp6]"=&f"(ftmp[6]), [ftmp7]"=&f"(ftmp[7]), - [tmp0]"=&r"(tmp[0]), - RESTRICT_ASM_ALL64 - RESTRICT_ASM_ADDRT - [addr0]"=&r"(addr[0]), + [ftmp8]"=&f"(ftmp[8]), [tmp0]"=&r"(tmp[0]), [pixels]"+&r"(pixels), [block]"+&r"(block) : [line_size]"r"((mips_reg)line_size) : "memory"