From: "Rémi Denis-Courmont" <remi@remlab.net>
To: ffmpeg-devel@ffmpeg.org
Subject: [FFmpeg-devel] [PATCH 3/9] lavc/riscv: require B or zba explicitly
Date: Mon, 22 Jul 2024 21:44:22 +0300
Message-ID: <20240722184431.40853-3-remi@remlab.net> (raw)
In-Reply-To: <20240722184431.40853-1-remi@remlab.net>
---
libavcodec/riscv/aacencdsp_rvv.S | 4 +--
libavcodec/riscv/aacpsdsp_rvv.S | 10 +++----
libavcodec/riscv/ac3dsp_rvv.S | 6 ++--
libavcodec/riscv/ac3dsp_rvvb.S | 2 +-
libavcodec/riscv/alacdsp_rvv.S | 6 ++--
libavcodec/riscv/audiodsp_rvv.S | 6 ++--
libavcodec/riscv/bswapdsp_rvb.S | 2 +-
libavcodec/riscv/bswapdsp_rvv.S | 2 +-
libavcodec/riscv/exrdsp_rvv.S | 2 +-
libavcodec/riscv/fixed_vtype.S | 48 ++++++++++++++++++++++++++++++
libavcodec/riscv/flacdsp_rvv.S | 42 +++++++++++++-------------
libavcodec/riscv/fmtconvert_rvv.S | 4 +--
libavcodec/riscv/h264_mc_chroma.S | 4 +--
libavcodec/riscv/h264idct_rvv.S | 2 +-
libavcodec/riscv/huffyuvdsp_rvv.S | 4 +--
libavcodec/riscv/jpeg2000dsp_rvv.S | 4 +--
libavcodec/riscv/llauddsp_rvv.S | 4 +--
libavcodec/riscv/lpc_rvv.S | 4 +--
libavcodec/riscv/opusdsp_rvv.S | 2 +-
libavcodec/riscv/rv40dsp_rvv.S | 4 +--
libavcodec/riscv/sbrdsp_rvv.S | 16 +++++-----
libavcodec/riscv/svqenc_rvv.S | 2 +-
libavcodec/riscv/takdsp_rvv.S | 8 ++---
libavcodec/riscv/utvideodsp_rvv.S | 4 +--
libavcodec/riscv/vc1dsp_rvv.S | 6 ++--
libavcodec/riscv/vorbisdsp_rvv.S | 2 +-
libavcodec/riscv/vp7dsp_rvv.S | 2 +-
libavcodec/riscv/vp8dsp_rvv.S | 4 +--
libavcodec/riscv/vp9_intra_rvi.S | 6 ++--
29 files changed, 129 insertions(+), 83 deletions(-)
create mode 100644 libavcodec/riscv/fixed_vtype.S
diff --git a/libavcodec/riscv/aacencdsp_rvv.S b/libavcodec/riscv/aacencdsp_rvv.S
index 21e66a77ae..05a603b6f6 100644
--- a/libavcodec/riscv/aacencdsp_rvv.S
+++ b/libavcodec/riscv/aacencdsp_rvv.S
@@ -21,7 +21,7 @@
#include "libavutil/riscv/asm.S"
-func ff_abs_pow34_rvv, zve32f
+func ff_abs_pow34_rvv, zve32f, zba
1:
vsetvli t0, a2, e32, m8, ta, ma
sub a2, a2, t0
@@ -38,7 +38,7 @@ func ff_abs_pow34_rvv, zve32f
ret
endfunc
-func ff_aac_quant_bands_rvv, zve32f
+func ff_aac_quant_bands_rvv, zve32f, zba
NOHWF fmv.w.x fa0, a6
NOHWF fmv.w.x fa1, a7
fcvt.s.w ft0, a5
diff --git a/libavcodec/riscv/aacpsdsp_rvv.S b/libavcodec/riscv/aacpsdsp_rvv.S
index 2d6858688a..72e2103c22 100644
--- a/libavcodec/riscv/aacpsdsp_rvv.S
+++ b/libavcodec/riscv/aacpsdsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_ps_add_squares_rvv, zve64f
+func ff_ps_add_squares_rvv, zve64f, zba
li t1, 32
1:
vsetvli t0, a2, e32, m4, ta, ma
@@ -39,7 +39,7 @@ func ff_ps_add_squares_rvv, zve64f
ret
endfunc
-func ff_ps_mul_pair_single_rvv, zve32f
+func ff_ps_mul_pair_single_rvv, zve32f, zba
1:
vsetvli t0, a3, e32, m4, ta, ma
vlseg2e32.v v24, (a1)
@@ -134,7 +134,7 @@ NOHWD flw fs\n, (4 * \n)(sp)
.purgem filter
endfunc
-func ff_ps_hybrid_analysis_ileave_rvv, zve32x /* no needs for zve32f here */
+func ff_ps_hybrid_analysis_ileave_rvv, zve32x /* no zve32f here */, zba
slli t0, a2, 5 + 1 + 2 // ctz(32 * 2 * 4)
sh2add a1, a2, a1
add a0, a0, t0
@@ -169,7 +169,7 @@ func ff_ps_hybrid_analysis_ileave_rvv, zve32x /* no needs for zve32f here */
ret
endfunc
-func ff_ps_hybrid_synthesis_deint_rvv, zve64x
+func ff_ps_hybrid_synthesis_deint_rvv, zve64x, zba
slli t0, a2, 5 + 1 + 2
sh2add a0, a2, a0
add a1, a1, t0
@@ -207,7 +207,7 @@ func ff_ps_hybrid_synthesis_deint_rvv, zve64x
ret
endfunc
-func ff_ps_stereo_interpolate_rvv, zve32f, zbb
+func ff_ps_stereo_interpolate_rvv, zve32f, b
vsetvli t0, zero, e32, m2, ta, ma
vid.v v24
flw ft0, (a2)
diff --git a/libavcodec/riscv/ac3dsp_rvv.S b/libavcodec/riscv/ac3dsp_rvv.S
index 1b5f67a9ec..c733733286 100644
--- a/libavcodec/riscv/ac3dsp_rvv.S
+++ b/libavcodec/riscv/ac3dsp_rvv.S
@@ -43,7 +43,7 @@ func ff_ac3_exponent_min_rvv, zve32x
ret
endfunc
-func ff_float_to_fixed24_rvv, zve32f
+func ff_float_to_fixed24_rvv, zve32f, zba
li t1, 1 << 24
fcvt.s.w f0, t1
1:
@@ -61,7 +61,7 @@ func ff_float_to_fixed24_rvv, zve32f
endfunc
#if __riscv_xlen >= 64
-func ff_sum_square_butterfly_int32_rvv, zve64x
+func ff_sum_square_butterfly_int32_rvv, zve64x, zba
vsetvli t0, zero, e64, m8, ta, ma
vmv.v.x v0, zero
vmv.v.x v8, zero
@@ -101,7 +101,7 @@ func ff_sum_square_butterfly_int32_rvv, zve64x
endfunc
#endif
-func ff_sum_square_butterfly_float_rvv, zve32f
+func ff_sum_square_butterfly_float_rvv, zve32f, zba
vsetvli t0, zero, e32, m8, ta, ma
vmv.v.x v0, zero
vmv.v.x v8, zero
diff --git a/libavcodec/riscv/ac3dsp_rvvb.S b/libavcodec/riscv/ac3dsp_rvvb.S
index 64766b56be..5bffb40bba 100644
--- a/libavcodec/riscv/ac3dsp_rvvb.S
+++ b/libavcodec/riscv/ac3dsp_rvvb.S
@@ -21,7 +21,7 @@
#include "config.h"
#include "libavutil/riscv/asm.S"
-func ff_extract_exponents_rvvb, zve32x, zvbb
+func ff_extract_exponents_rvvb, zve32x, zvbb, zba
1:
vsetvli t0, a2, e32, m8, ta, ma
vle32.v v8, (a1)
diff --git a/libavcodec/riscv/alacdsp_rvv.S b/libavcodec/riscv/alacdsp_rvv.S
index 8efb04e0c8..19714bd6e3 100644
--- a/libavcodec/riscv/alacdsp_rvv.S
+++ b/libavcodec/riscv/alacdsp_rvv.S
@@ -21,7 +21,7 @@
#include "libavutil/riscv/asm.S"
#if (__riscv_xlen == 64)
-func ff_alac_decorrelate_stereo_rvv, zve32x
+func ff_alac_decorrelate_stereo_rvv, zve32x, zba
ld a4, 8(a0)
ld a0, 0(a0)
1:
@@ -43,7 +43,7 @@ func ff_alac_decorrelate_stereo_rvv, zve32x
ret
endfunc
-func ff_alac_append_extra_bits_mono_rvv, zve32x
+func ff_alac_append_extra_bits_mono_rvv, zve32x, zba
ld a0, (a0)
ld a1, (a1)
1:
@@ -61,7 +61,7 @@ func ff_alac_append_extra_bits_mono_rvv, zve32x
ret
endfunc
-func ff_alac_append_extra_bits_stereo_rvv, zve32x
+func ff_alac_append_extra_bits_stereo_rvv, zve32x, zba
ld a6, 8(a0)
ld a0, (a0)
ld a7, 8(a1)
diff --git a/libavcodec/riscv/audiodsp_rvv.S b/libavcodec/riscv/audiodsp_rvv.S
index f0b23bab5e..b7134de523 100644
--- a/libavcodec/riscv/audiodsp_rvv.S
+++ b/libavcodec/riscv/audiodsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_scalarproduct_int16_rvv, zve32x
+func ff_scalarproduct_int16_rvv, zve32x, zba
vsetvli t0, zero, e32, m8, ta, ma
vmv.v.x v8, zero
vmv.s.x v0, zero
@@ -40,7 +40,7 @@ func ff_scalarproduct_int16_rvv, zve32x
ret
endfunc
-func ff_vector_clip_int32_rvv, zve32x
+func ff_vector_clip_int32_rvv, zve32x, zba
1:
vsetvli t0, a4, e32, m8, ta, ma
vle32.v v8, (a1)
@@ -55,7 +55,7 @@ func ff_vector_clip_int32_rvv, zve32x
ret
endfunc
-func ff_vector_clipf_rvv, zve32f
+func ff_vector_clipf_rvv, zve32f, zba
NOHWF fmv.w.x fa0, a3
NOHWF fmv.w.x fa1, a4
1:
diff --git a/libavcodec/riscv/bswapdsp_rvb.S b/libavcodec/riscv/bswapdsp_rvb.S
index 0786bd3f36..17cfd5d7ef 100644
--- a/libavcodec/riscv/bswapdsp_rvb.S
+++ b/libavcodec/riscv/bswapdsp_rvb.S
@@ -23,7 +23,7 @@
#include "libavutil/riscv/bswap_rvb.S"
#if (__riscv_xlen >= 64)
-func ff_bswap32_buf_rvb, zbb
+func ff_bswap32_buf_rvb, zba, zbb
bswap32_rvb a0, a1, a2
endfunc
#endif
diff --git a/libavcodec/riscv/bswapdsp_rvv.S b/libavcodec/riscv/bswapdsp_rvv.S
index b37fe26255..14484a772d 100644
--- a/libavcodec/riscv/bswapdsp_rvv.S
+++ b/libavcodec/riscv/bswapdsp_rvv.S
@@ -21,7 +21,7 @@
#include "config.h"
#include "libavutil/riscv/asm.S"
-func ff_bswap16_buf_rvv, zve32x
+func ff_bswap16_buf_rvv, zve32x, zba
1:
vsetvli t0, a2, e16, m8, ta, ma
vle16.v v8, (a1)
diff --git a/libavcodec/riscv/exrdsp_rvv.S b/libavcodec/riscv/exrdsp_rvv.S
index f4a35f58ff..c1d7dfcb86 100644
--- a/libavcodec/riscv/exrdsp_rvv.S
+++ b/libavcodec/riscv/exrdsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_reorder_pixels_rvv, zve32x
+func ff_reorder_pixels_rvv, zve32x, zba
srai a2, a2, 1
add t1, a1, a2
1:
diff --git a/libavcodec/riscv/fixed_vtype.S b/libavcodec/riscv/fixed_vtype.S
new file mode 100644
index 0000000000..7aac70bda8
--- /dev/null
+++ b/libavcodec/riscv/fixed_vtype.S
@@ -0,0 +1,48 @@
+/*
+ * Copyright © 2024 Rémi Denis-Courmont.
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#define VILL -0x8000
+
+ .data
+ff_vtype_e8:
+ff_vtype_e8_1:
+ .half VILL
+ff_vtype_e8_2:
+ .half VILL
+ff_vtype_e8_4:
+ .half VILL
+ff_vtype_e8_8:
+ .half VILL
+ff_vtype_e8_16:
+ .half VILL
+ff_vtype_e8_32:
+ .half VILL
+
+ff_vtype_e16:
+ff_vtype_e16_1:
+ .half VILL
+ff_vtype_e16_2:
+ .half VILL
+ff_vtype_e16_4:
+ .half VILL
+ff_vtype_e16_8:
+ .half VILL
+ff_vtype_e16_16:
+ .half VILL
diff --git a/libavcodec/riscv/flacdsp_rvv.S b/libavcodec/riscv/flacdsp_rvv.S
index 1724aee9d7..69505c694a 100644
--- a/libavcodec/riscv/flacdsp_rvv.S
+++ b/libavcodec/riscv/flacdsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_flac_lpc16_rvv, zve32x, zbb
+func ff_flac_lpc16_rvv, zve32x, b
vtype_vli t0, a2, t2, e32, ta, ma
vsetvl zero, a2, t0
vle32.v v8, (a1)
@@ -45,7 +45,7 @@ func ff_flac_lpc16_rvv, zve32x, zbb
endfunc
#if (__riscv_xlen == 64)
-func ff_flac_lpc32_rvv, zve64x
+func ff_flac_lpc32_rvv, zve64x, zba
addi t2, a2, -16
ble t2, zero, ff_flac_lpc32_rvv_simple
vsetivli zero, 1, e64, m1, ta, ma
@@ -76,7 +76,7 @@ func ff_flac_lpc32_rvv, zve64x
ret
endfunc
-func ff_flac_lpc32_rvv_simple, zve64x, zbb
+func ff_flac_lpc32_rvv_simple, zve64x, b
vtype_vli t3, a2, t1, e64, ta, ma
vntypei t2, t3
vsetvl zero, a2, t3 // e64
@@ -104,7 +104,7 @@ func ff_flac_lpc32_rvv_simple, zve64x, zbb
ret
endfunc
-func ff_flac_lpc33_rvv, zve64x, zbb
+func ff_flac_lpc33_rvv, zve64x, b
vtype_vli t0, a3, t1, e64, ta, ma
vsetvl zero, a3, t0
vmv.s.x v0, zero
@@ -132,7 +132,7 @@ func ff_flac_lpc33_rvv, zve64x, zbb
endfunc
#endif
-func ff_flac_wasted32_rvv, zve32x
+func ff_flac_wasted32_rvv, zve32x, zba
1:
vsetvli t0, a2, e32, m8, ta, ma
vle32.v v8, (a0)
@@ -145,7 +145,7 @@ func ff_flac_wasted32_rvv, zve32x
ret
endfunc
-func ff_flac_wasted33_rvv, zve64x
+func ff_flac_wasted33_rvv, zve64x, zba
srli t0, a2, 5
li t1, 1
bnez t0, 2f
@@ -177,7 +177,7 @@ func ff_flac_wasted33_rvv, zve64x
endfunc
#if (__riscv_xlen == 64)
-func ff_flac_decorrelate_indep2_16_rvv, zve32x
+func ff_flac_decorrelate_indep2_16_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld a1, (a1)
@@ -200,7 +200,7 @@ func ff_flac_decorrelate_indep2_16_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_indep4_16_rvv, zve32x
+func ff_flac_decorrelate_indep4_16_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld t1, 16(a1)
@@ -233,7 +233,7 @@ func ff_flac_decorrelate_indep4_16_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_indep6_16_rvv, zve32x
+func ff_flac_decorrelate_indep6_16_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld t1, 16(a1)
@@ -278,7 +278,7 @@ func ff_flac_decorrelate_indep6_16_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_indep8_16_rvv, zve32x
+func ff_flac_decorrelate_indep8_16_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld t1, 16(a1)
@@ -332,9 +332,7 @@ func ff_flac_decorrelate_indep8_16_rvv, zve32x
ret
endfunc
-
-
-func ff_flac_decorrelate_ls_16_rvv, zve32x
+func ff_flac_decorrelate_ls_16_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld a1, (a1)
@@ -358,7 +356,7 @@ func ff_flac_decorrelate_ls_16_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_rs_16_rvv, zve32x
+func ff_flac_decorrelate_rs_16_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld a1, (a1)
@@ -382,7 +380,7 @@ func ff_flac_decorrelate_rs_16_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_ms_16_rvv, zve32x
+func ff_flac_decorrelate_ms_16_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld a1, (a1)
@@ -408,7 +406,7 @@ func ff_flac_decorrelate_ms_16_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_indep2_32_rvv, zve32x
+func ff_flac_decorrelate_indep2_32_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld a1, (a1)
@@ -428,7 +426,7 @@ func ff_flac_decorrelate_indep2_32_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_indep4_32_rvv, zve32x
+func ff_flac_decorrelate_indep4_32_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld t1, 16(a1)
@@ -457,7 +455,7 @@ func ff_flac_decorrelate_indep4_32_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_indep6_32_rvv, zve32x
+func ff_flac_decorrelate_indep6_32_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld t1, 16(a1)
@@ -495,7 +493,7 @@ func ff_flac_decorrelate_indep6_32_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_indep8_32_rvv, zve32x
+func ff_flac_decorrelate_indep8_32_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld t1, 16(a1)
@@ -540,7 +538,7 @@ func ff_flac_decorrelate_indep8_32_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_ls_32_rvv, zve32x
+func ff_flac_decorrelate_ls_32_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld a1, (a1)
@@ -561,7 +559,7 @@ func ff_flac_decorrelate_ls_32_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_rs_32_rvv, zve32x
+func ff_flac_decorrelate_rs_32_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld a1, (a1)
@@ -582,7 +580,7 @@ func ff_flac_decorrelate_rs_32_rvv, zve32x
ret
endfunc
-func ff_flac_decorrelate_ms_32_rvv, zve32x
+func ff_flac_decorrelate_ms_32_rvv, zve32x, zba
ld a0, (a0)
ld a2, 8(a1)
ld a1, (a1)
diff --git a/libavcodec/riscv/fmtconvert_rvv.S b/libavcodec/riscv/fmtconvert_rvv.S
index d0e2f106d5..05cd3b38a5 100644
--- a/libavcodec/riscv/fmtconvert_rvv.S
+++ b/libavcodec/riscv/fmtconvert_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_int32_to_float_fmul_scalar_rvv, zve32f
+func ff_int32_to_float_fmul_scalar_rvv, zve32f, zba
NOHWF fmv.w.x fa0, a2
NOHWF mv a2, a3
1:
@@ -37,7 +37,7 @@ NOHWF mv a2, a3
ret
endfunc
-func ff_int32_to_float_fmul_array8_rvv, zve32f
+func ff_int32_to_float_fmul_array8_rvv, zve32f, zba
srai a4, a4, 3
1: vsetvli t0, a4, e32, m1, ta, ma
diff --git a/libavcodec/riscv/h264_mc_chroma.S b/libavcodec/riscv/h264_mc_chroma.S
index ce99bda44d..b6c0e1c635 100644
--- a/libavcodec/riscv/h264_mc_chroma.S
+++ b/libavcodec/riscv/h264_mc_chroma.S
@@ -325,7 +325,7 @@
ret
.endm
-func h264_put_chroma_mc_rvv, zve32x
+func h264_put_chroma_mc_rvv, zve32x, zba
11:
li a7, 3
blt a3, a7, 12f
@@ -334,7 +334,7 @@ func h264_put_chroma_mc_rvv, zve32x
do_chroma_mc put 0
endfunc
-func h264_avg_chroma_mc_rvv, zve32x
+func h264_avg_chroma_mc_rvv, zve32x, zba
21:
li a7, 3
blt a3, a7, 22f
diff --git a/libavcodec/riscv/h264idct_rvv.S b/libavcodec/riscv/h264idct_rvv.S
index c74ea18c19..514c849bce 100644
--- a/libavcodec/riscv/h264idct_rvv.S
+++ b/libavcodec/riscv/h264idct_rvv.S
@@ -298,7 +298,7 @@ func ff_h264_idct8_add_8_rvv, zve32x
ret
endfunc
-func ff_h264_idct8_add_16_rvv, zve32x
+func ff_h264_idct8_add_16_rvv, zve32x, zba
csrwi vxrm, 0
.Lidct8_add_16_rvv:
li a4, 8
diff --git a/libavcodec/riscv/huffyuvdsp_rvv.S b/libavcodec/riscv/huffyuvdsp_rvv.S
index d334f5c6d0..54d1d94059 100644
--- a/libavcodec/riscv/huffyuvdsp_rvv.S
+++ b/libavcodec/riscv/huffyuvdsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_add_int16_rvv, zve32x
+func ff_add_int16_rvv, zve32x, zba
1:
vsetvli t0, a3, e16, m8, ta, ma
vle16.v v16, (a0)
@@ -36,7 +36,7 @@ func ff_add_int16_rvv, zve32x
ret
endfunc
-func ff_add_hfyu_left_pred_bgr32_rvv, zve32x, zbb
+func ff_add_hfyu_left_pred_bgr32_rvv, zve32x, b
vtype_ivli t1, 4, e8, ta, ma
li t0, 4
vsetvl zero, t0, t1
diff --git a/libavcodec/riscv/jpeg2000dsp_rvv.S b/libavcodec/riscv/jpeg2000dsp_rvv.S
index 10efe6b0db..77c6fd2d32 100644
--- a/libavcodec/riscv/jpeg2000dsp_rvv.S
+++ b/libavcodec/riscv/jpeg2000dsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_ict_float_rvv, zve32f
+func ff_ict_float_rvv, zve32f, zba
lla t0, ff_jpeg2000_f_ict_params
flw ft0, 0(t0)
flw ft1, 4(t0)
@@ -48,7 +48,7 @@ func ff_ict_float_rvv, zve32f
ret
endfunc
-func ff_rct_int_rvv, zve32x
+func ff_rct_int_rvv, zve32x, zba
1:
vsetvli t0, a3, e32, m8, ta, ma
vle32.v v16, (a1)
diff --git a/libavcodec/riscv/llauddsp_rvv.S b/libavcodec/riscv/llauddsp_rvv.S
index 5569864832..6af2e6a882 100644
--- a/libavcodec/riscv/llauddsp_rvv.S
+++ b/libavcodec/riscv/llauddsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_scalarproduct_and_madd_int16_rvv, zve32x
+func ff_scalarproduct_and_madd_int16_rvv, zve32x, zba
vsetvli t0, zero, e32, m8, ta, ma
vmv.v.x v0, zero
1:
@@ -44,7 +44,7 @@ func ff_scalarproduct_and_madd_int16_rvv, zve32x
ret
endfunc
-func ff_scalarproduct_and_madd_int32_rvv, zve32x
+func ff_scalarproduct_and_madd_int32_rvv, zve32x, zba
vsetvli t0, zero, e32, m8, ta, ma
vmv.v.x v0, zero
1:
diff --git a/libavcodec/riscv/lpc_rvv.S b/libavcodec/riscv/lpc_rvv.S
index fe80305d9a..8fd1ddbbf0 100644
--- a/libavcodec/riscv/lpc_rvv.S
+++ b/libavcodec/riscv/lpc_rvv.S
@@ -21,7 +21,7 @@
#include "libavutil/riscv/asm.S"
#if __riscv_xlen >= 64
-func ff_lpc_apply_welch_window_rvv, zve64d
+func ff_lpc_apply_welch_window_rvv, zve64d, zba
vsetvli t0, zero, e64, m8, ta, ma
vid.v v0
addi t2, a1, -1
@@ -86,7 +86,7 @@ func ff_lpc_apply_welch_window_rvv, zve64d
ret
endfunc
-func ff_lpc_compute_autocorr_rvv, zve64d, zbb
+func ff_lpc_compute_autocorr_rvv, zve64d, b
vtype_vli t1, a2, t2, e64, ta, ma, 1
addi a2, a2, 1
li t0, 1
diff --git a/libavcodec/riscv/opusdsp_rvv.S b/libavcodec/riscv/opusdsp_rvv.S
index 42d845a370..389dc744f5 100644
--- a/libavcodec/riscv/opusdsp_rvv.S
+++ b/libavcodec/riscv/opusdsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_opus_postfilter_rvv, zve32f, zbb
+func ff_opus_postfilter_rvv, zve32f, b
flw fa0, 0(a2) // g0
slli t1, a1, 2
flw fa1, 4(a2) // g1
diff --git a/libavcodec/riscv/rv40dsp_rvv.S b/libavcodec/riscv/rv40dsp_rvv.S
index e49345ef70..53d3d1d6f9 100644
--- a/libavcodec/riscv/rv40dsp_rvv.S
+++ b/libavcodec/riscv/rv40dsp_rvv.S
@@ -332,7 +332,7 @@
ret
.endm
-func ff_put_rv40_chroma_mc_rvv, zve32x
+func ff_put_rv40_chroma_mc_rvv, zve32x, zba
11:
li a7, 3
blt a3, a7, 12f
@@ -341,7 +341,7 @@ func ff_put_rv40_chroma_mc_rvv, zve32x
do_chroma_mc put 0
endfunc
-func ff_avg_rv40_chroma_mc_rvv, zve32x
+func ff_avg_rv40_chroma_mc_rvv, zve32x, zba
21:
li a7, 3
blt a3, a7, 22f
diff --git a/libavcodec/riscv/sbrdsp_rvv.S b/libavcodec/riscv/sbrdsp_rvv.S
index 331b88022c..7c90a8addf 100644
--- a/libavcodec/riscv/sbrdsp_rvv.S
+++ b/libavcodec/riscv/sbrdsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_sbr_sum64x5_rvv, zve32f
+func ff_sbr_sum64x5_rvv, zve32f, zba
li a5, 64
addi a1, a0, 64 * 4
addi a2, a0, 128 * 4
@@ -49,7 +49,7 @@ func ff_sbr_sum64x5_rvv, zve32f
ret
endfunc
-func ff_sbr_sum_square_rvv, zve32f
+func ff_sbr_sum_square_rvv, zve32f, zba
vsetvli t0, zero, e32, m8, ta, ma
slli a1, a1, 1
vmv.v.x v8, zero
@@ -157,7 +157,7 @@ func ff_sbr_autocorrelate_rvv, zve32f
ret
endfunc
-func ff_sbr_hf_gen_rvv, zve32f
+func ff_sbr_hf_gen_rvv, zve32f, zba
NOHWF fmv.w.x fa0, a4
NOHWF mv a4, a5
NOHWF mv a5, a6
@@ -207,7 +207,7 @@ NOHWF mv a5, a6
ret
endfunc
-func ff_sbr_hf_g_filt_rvv, zve32f
+func ff_sbr_hf_g_filt_rvv, zve32f, zba
li t1, 40 * 2 * 4
sh3add a1, a4, a1
1:
@@ -272,16 +272,16 @@ endfunc
ret
.endm
-func ff_sbr_hf_apply_noise_0_rvv, zve32f, zbb
+func ff_sbr_hf_apply_noise_0_rvv, zve32f, b
hf_apply_noise 0
endfunc
-func ff_sbr_hf_apply_noise_3_rvv, zve32f, zbb
+func ff_sbr_hf_apply_noise_3_rvv, zve32f, b
not a4, a4 // invert parity of kx
// fall through
endfunc
-func ff_sbr_hf_apply_noise_1_rvv, zve32f, zbb
+func ff_sbr_hf_apply_noise_1_rvv, zve32f, b
vsetvli t0, zero, e32, m4, ta, ma
vid.v v4
vxor.vx v4, v4, a4
@@ -289,6 +289,6 @@ func ff_sbr_hf_apply_noise_1_rvv, zve32f, zbb
hf_apply_noise 1
endfunc
-func ff_sbr_hf_apply_noise_2_rvv, zve32f, zbb
+func ff_sbr_hf_apply_noise_2_rvv, zve32f, b
hf_apply_noise 2
endfunc
diff --git a/libavcodec/riscv/svqenc_rvv.S b/libavcodec/riscv/svqenc_rvv.S
index cfc27154dd..8b7a8b0400 100644
--- a/libavcodec/riscv/svqenc_rvv.S
+++ b/libavcodec/riscv/svqenc_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_ssd_int8_vs_int16_rvv, zve32x
+func ff_ssd_int8_vs_int16_rvv, zve32x, zba
vsetvli t0, zero, e32, m8, ta, ma
vmv.v.x v24, zero
1:
diff --git a/libavcodec/riscv/takdsp_rvv.S b/libavcodec/riscv/takdsp_rvv.S
index fa942a3be6..f3a230ccec 100644
--- a/libavcodec/riscv/takdsp_rvv.S
+++ b/libavcodec/riscv/takdsp_rvv.S
@@ -21,7 +21,7 @@
#include "libavutil/riscv/asm.S"
-func ff_decorrelate_ls_rvv, zve32x
+func ff_decorrelate_ls_rvv, zve32x, zba
1:
vsetvli t0, a2, e32, m8, ta, ma
sub a2, a2, t0
@@ -35,7 +35,7 @@ func ff_decorrelate_ls_rvv, zve32x
ret
endfunc
-func ff_decorrelate_sr_rvv, zve32x
+func ff_decorrelate_sr_rvv, zve32x, zba
1:
vsetvli t0, a2, e32, m8, ta, ma
vle32.v v0, (a0)
@@ -49,7 +49,7 @@ func ff_decorrelate_sr_rvv, zve32x
ret
endfunc
-func ff_decorrelate_sm_rvv, zve32x
+func ff_decorrelate_sm_rvv, zve32x, zba
1:
vsetvli t0, a2, e32, m8, ta, ma
vle32.v v8, (a1)
@@ -67,7 +67,7 @@ func ff_decorrelate_sm_rvv, zve32x
ret
endfunc
-func ff_decorrelate_sf_rvv, zve32x
+func ff_decorrelate_sf_rvv, zve32x, zba
csrwi vxrm, 0
1:
vsetvli t0, a2, e32, m8, ta, ma
diff --git a/libavcodec/riscv/utvideodsp_rvv.S b/libavcodec/riscv/utvideodsp_rvv.S
index fa70d0eb34..5e833eeb3c 100644
--- a/libavcodec/riscv/utvideodsp_rvv.S
+++ b/libavcodec/riscv/utvideodsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_restore_rgb_planes_rvv, zve32x
+func ff_restore_rgb_planes_rvv, zve32x, zba
li t1, -0x80
sub a3, a3, a6
sub a4, a4, a6
@@ -52,7 +52,7 @@ func ff_restore_rgb_planes_rvv, zve32x
ret
endfunc
-func ff_restore_rgb_planes10_rvv, zve32x
+func ff_restore_rgb_planes10_rvv, zve32x, zba
li t1, -0x200
li t2, 0x3FF
sub a3, a3, a6
diff --git a/libavcodec/riscv/vc1dsp_rvv.S b/libavcodec/riscv/vc1dsp_rvv.S
index d8b62579aa..aede87ccc0 100644
--- a/libavcodec/riscv/vc1dsp_rvv.S
+++ b/libavcodec/riscv/vc1dsp_rvv.S
@@ -21,7 +21,7 @@
#include "libavutil/riscv/asm.S"
-func ff_vc1_inv_trans_8x8_dc_rvv, zve64x
+func ff_vc1_inv_trans_8x8_dc_rvv, zve64x, zba
lh t2, (a2)
vsetivli zero, 8, e8, mf2, ta, ma
vlse64.v v0, (a0), a1
@@ -43,7 +43,7 @@ func ff_vc1_inv_trans_8x8_dc_rvv, zve64x
ret
endfunc
-func ff_vc1_inv_trans_4x8_dc_rvv, zve32x
+func ff_vc1_inv_trans_4x8_dc_rvv, zve32x, zba
lh t2, (a2)
vsetivli zero, 8, e8, mf2, ta, ma
vlse32.v v0, (a0), a1
@@ -67,7 +67,7 @@ func ff_vc1_inv_trans_4x8_dc_rvv, zve32x
ret
endfunc
-func ff_vc1_inv_trans_8x4_dc_rvv, zve64x
+func ff_vc1_inv_trans_8x4_dc_rvv, zve64x, zba
lh t2, (a2)
vsetivli zero, 4, e8, mf4, ta, ma
vlse64.v v0, (a0), a1
diff --git a/libavcodec/riscv/vorbisdsp_rvv.S b/libavcodec/riscv/vorbisdsp_rvv.S
index 81a6c62a65..d136188d2e 100644
--- a/libavcodec/riscv/vorbisdsp_rvv.S
+++ b/libavcodec/riscv/vorbisdsp_rvv.S
@@ -20,7 +20,7 @@
#include "libavutil/riscv/asm.S"
-func ff_vorbis_inverse_coupling_rvv, zve32f
+func ff_vorbis_inverse_coupling_rvv, zve32f, zba
fmv.w.x ft0, zero
1:
vsetvli t0, a2, e32, m4, ta, ma
diff --git a/libavcodec/riscv/vp7dsp_rvv.S b/libavcodec/riscv/vp7dsp_rvv.S
index 856b0e8c96..bfcc220273 100644
--- a/libavcodec/riscv/vp7dsp_rvv.S
+++ b/libavcodec/riscv/vp7dsp_rvv.S
@@ -21,7 +21,7 @@
#include "libavutil/riscv/asm.S"
#if __riscv_xlen >= 64
-func ff_vp7_luma_dc_wht_rvv, zve32x
+func ff_vp7_luma_dc_wht_rvv, zve32x, zba
li a2, 4 * 16 * 2
li a7, 16 * 2
jal t0, 1f
diff --git a/libavcodec/riscv/vp8dsp_rvv.S b/libavcodec/riscv/vp8dsp_rvv.S
index d366748a0a..4675f4c76b 100644
--- a/libavcodec/riscv/vp8dsp_rvv.S
+++ b/libavcodec/riscv/vp8dsp_rvv.S
@@ -440,7 +440,7 @@ endconst
.endm
.macro epel len size type
-func ff_put_vp8_epel\len\()_\type\()\size\()_rvv, zve32x
+func ff_put_vp8_epel\len\()_\type\()\size\()_rvv, zve32x, zba
epel_filter \size \type t
vsetvlstatic8 \len
1:
@@ -455,7 +455,7 @@ endfunc
.endm
.macro epel_hv len hsize vsize
-func ff_put_vp8_epel\len\()_h\hsize\()v\vsize\()_rvv, zve32x
+func ff_put_vp8_epel\len\()_h\hsize\()v\vsize\()_rvv, zve32x, zba
#if __riscv_xlen == 64
addi sp, sp, -48
.irp n,0,1,2,3,4,5
diff --git a/libavcodec/riscv/vp9_intra_rvi.S b/libavcodec/riscv/vp9_intra_rvi.S
index 16b6bdb25a..dadd4be194 100644
--- a/libavcodec/riscv/vp9_intra_rvi.S
+++ b/libavcodec/riscv/vp9_intra_rvi.S
@@ -21,7 +21,7 @@
#include "libavutil/riscv/asm.S"
#if __riscv_xlen >= 64
-func ff_v_32x32_rvi
+func ff_v_32x32_rvi, zba
ld t0, (a3)
ld t1, 8(a3)
ld t2, 16(a3)
@@ -42,7 +42,7 @@ func ff_v_32x32_rvi
ret
endfunc
-func ff_v_16x16_rvi
+func ff_v_16x16_rvi, zba
ld t0, (a3)
ld t1, 8(a3)
.rept 8
@@ -57,7 +57,7 @@ func ff_v_16x16_rvi
ret
endfunc
-func ff_v_8x8_rvi
+func ff_v_8x8_rvi, zba
ld t0, (a3)
.rept 4
add a7, a0, a1
--
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".
next prev parent reply other threads:[~2024-07-22 18:44 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-22 18:44 [FFmpeg-devel] [PATCH 1/9] lavu/riscv: allow any number of extensions Rémi Denis-Courmont
2024-07-22 18:44 ` [FFmpeg-devel] [PATCH 2/9] lavu/riscv: grok B as an extension Rémi Denis-Courmont
2024-07-22 18:44 ` Rémi Denis-Courmont [this message]
2024-07-22 18:44 ` [FFmpeg-devel] [PATCH 4/9] lavfi/riscv: require B or zba explicitly Rémi Denis-Courmont
2024-07-22 18:44 ` [FFmpeg-devel] [PATCH 5/9] sws/riscv: " Rémi Denis-Courmont
2024-07-22 18:44 ` [FFmpeg-devel] [PATCH 6/9] lavu/riscv: " Rémi Denis-Courmont
2024-07-22 18:44 ` [FFmpeg-devel] [PATCH 7/9] lavu/riscv: remove bespoke SH{1, 2, 3}ADD assembler Rémi Denis-Courmont
2024-07-22 18:44 ` [FFmpeg-devel] [PATCH 8/9] lavu/riscv: add CPU flag for B bit manipulations Rémi Denis-Courmont
2024-07-22 18:44 ` [FFmpeg-devel] [PATCH 9/9] lavc/h264dsp: use RISC-V B extension Rémi Denis-Courmont
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240722184431.40853-3-remi@remlab.net \
--to=remi@remlab.net \
--cc=ffmpeg-devel@ffmpeg.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
This inbox may be cloned and mirrored by anyone:
git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git
# If you have public-inbox 1.1+ installed, you may
# initialize and index your mirror using the following commands:
public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \
ffmpegdev@gitmailbox.com
public-inbox-index ffmpegdev
Example config snippet for mirrors.
AGPL code for this site: git clone https://public-inbox.org/public-inbox.git