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 39CC14B984 for ; Tue, 2 Jul 2024 17:14:05 +0000 (UTC) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 44D3768D936; Tue, 2 Jul 2024 20:13:50 +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 CEFC568D903 for ; Tue, 2 Jul 2024 20:13:37 +0300 (EEST) Received: from basile.remlab.net (localhost [IPv6:::1]) by ursule.remlab.net (Postfix) with ESMTP id 3EB2FC018B for ; Tue, 2 Jul 2024 20:13:37 +0300 (EEST) From: =?UTF-8?q?R=C3=A9mi=20Denis-Courmont?= To: ffmpeg-devel@ffmpeg.org Date: Tue, 2 Jul 2024 20:13:35 +0300 Message-ID: <20240702171336.26390-3-remi@remlab.net> X-Mailer: git-send-email 2.45.2 In-Reply-To: <20240702171336.26390-1-remi@remlab.net> References: <20240702171336.26390-1-remi@remlab.net> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 3/3] lavc/h264dsp: R-V V 8-bit h264_idct8_add4 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: --- libavcodec/riscv/h264dsp_init.c | 4 ++ libavcodec/riscv/h264idct_rvv.S | 70 +++++++++++++++++++++++++++++++++ 2 files changed, 74 insertions(+) diff --git a/libavcodec/riscv/h264dsp_init.c b/libavcodec/riscv/h264dsp_init.c index 7badb86e87..5cf75f280e 100644 --- a/libavcodec/riscv/h264dsp_init.c +++ b/libavcodec/riscv/h264dsp_init.c @@ -40,6 +40,9 @@ void ff_h264_idct_add16_8_rvv(uint8_t *dst, const int *blockoffset, void ff_h264_idct_add16intra_8_rvv(uint8_t *dst, const int *blockoffset, int16_t *block, int stride, const uint8_t nnzc[5 * 8]); +void ff_h264_idct8_add4_8_rvv(uint8_t *dst, const int *blockoffset, + int16_t *block, int stride, + const uint8_t nnzc[5 * 8]); extern int ff_startcode_find_candidate_rvb(const uint8_t *, int); extern int ff_startcode_find_candidate_rvv(const uint8_t *, int); @@ -62,6 +65,7 @@ av_cold void ff_h264dsp_init_riscv(H264DSPContext *dsp, const int bit_depth, dsp->h264_idct_add16 = ff_h264_idct_add16_8_rvv; dsp->h264_idct_add16intra = ff_h264_idct_add16intra_8_rvv; + dsp->h264_idct8_add4 = ff_h264_idct8_add4_8_rvv; } dsp->startcode_find_candidate = ff_startcode_find_candidate_rvv; } diff --git a/libavcodec/riscv/h264idct_rvv.S b/libavcodec/riscv/h264idct_rvv.S index 42a29ba336..7422942717 100644 --- a/libavcodec/riscv/h264idct_rvv.S +++ b/libavcodec/riscv/h264idct_rvv.S @@ -170,5 +170,75 @@ func ff_h264_idct_add16intra_\depth\()_rvv, zve32x addi sp, sp, 80 ret endfunc + +func ff_h264_idct8_add4_\depth\()_rvv, zve32x + addi sp, sp, -80 + lla t0, ff_h264_scan8 + sd s0, (sp) + li t1, 4 * 32 << (\depth > 8) + mv s0, sp + li t2, 4 + sd ra, 8(sp) + sd s1, 16(sp) + sd s2, 24(sp) + sd s3, 32(sp) + sd s4, 40(sp) + sd s5, 48(sp) + sd s6, 56(sp) + sd s7, 64(sp) + vsetivli zero, 4, e8, mf4, ta, ma + vlse8.v v8, (t0), t2 + vlse16.v v16, (a2), t1 + vluxei8.v v12, (a4), v8 +.if \depth == 8 + vsetvli zero, zero, e16, mf2, ta, ma +.else + vsetvli zero, zero, e32, m1, ta, ma +.endif + vmsne.vi v1, v16, 0 + vsetvli zero, zero, e8, mf4, ta, ma + vmseq.vi v2, v12, 1 + vmsne.vi v0, v12, 0 + vmand.mm v1, v1, v2 + vmv.x.s s2, v0 + vmv.x.s s3, v1 + li s1, 4 + mv s4, a0 + mv s5, a1 + mv s6, a2 + mv s7, a3 +1: + andi t0, s2, 1 + addi s1, s1, -1 + srli s2, s2, 1 + beqz t0, 3f # if (nnz) + lw t2, (s5) # block_offset[i] + andi t1, s3, 1 + mv a1, s6 + mv a2, s7 + add a0, s4, t2 + beqz t1, 2f # if (nnz == 1 && block[i * 16]) + call ff_h264_idct8_dc_add_\depth\()_c + j 3f +2: + call ff_h264_idct8_add_\depth\()_c +3: + srli s3, s3, 1 + addi s5, s5, 4 * 4 + addi s6, s6, 4 * 16 * 2 << (\depth > 8) + bnez s1, 1b + + ld s7, 64(sp) + ld s6, 56(sp) + ld s5, 48(sp) + ld s4, 40(sp) + ld s3, 32(sp) + ld s2, 24(sp) + ld s1, 16(sp) + ld ra, 8(sp) + ld s0, 0(sp) + addi sp, sp, 80 + ret +endfunc .endr #endif -- 2.45.2 _______________________________________________ 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".