2 * Alpha optimized DSP utils
3 * Copyright (c) 2002 Falk Hueffner <falk@debian.org>
5 * This file is part of Libav.
7 * Libav is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Lesser General Public
9 * License as published by the Free Software Foundation; either
10 * version 2.1 of the License, or (at your option) any later version.
12 * Libav is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15 * Lesser General Public License for more details.
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with Libav; if not, write to the Free Software
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
23 * These functions are scheduled for pca56. They should work
24 * reasonably on ev6, though.
29 /* Some nicer register names. */
34 /* Danger: these overlap with the argument list and the return value */
45 /************************************************************************
46 * void put_pixels_axp_asm(uint8_t *block, const uint8_t *pixels,
47 * int line_size, int h)
50 .globl put_pixels_axp_asm
51 .ent put_pixels_axp_asm
135 .end put_pixels_axp_asm
137 /************************************************************************
138 * void put_pixels_clamped_mvi_asm(const DCTELEM *block, uint8_t *pixels,
142 .globl put_pixels_clamped_mvi_asm
143 .ent put_pixels_clamped_mvi_asm
144 put_pixels_clamped_mvi_asm:
149 lda t9, 8 # loop counter
150 zap t8, 0xaa, t8 # 00ff00ff00ff00ff
185 .end put_pixels_clamped_mvi_asm
187 /************************************************************************
188 * void add_pixels_clamped_mvi_asm(const DCTELEM *block, uint8_t *pixels,
192 .globl add_pixels_clamped_mvi_asm
193 .ent add_pixels_clamped_mvi_asm
194 add_pixels_clamped_mvi_asm:
204 xor tg, t0, tg # 0x8000800080008000
205 zap t1, 0xaa, tf # 0x00ff00ff00ff00ff
208 1: ldl t1, 0(a1) # pix0 (try to hit cache line soon)
210 addq a1, a2, te # pixels += line_size
211 ldq t0, 0(a0) # shorts0
213 ldl t7, 0(te) # pix2 (try to hit cache line soon)
215 ldq t3, 8(a0) # shorts1
216 ldq t6, 16(a0) # shorts2
218 ldq t9, 24(a0) # shorts3
219 unpkbw t1, t1 # 0 0 (quarter/op no.)
226 addq t0, t1, t0 # 0 3
231 maxsw4 t0, zero, t0 # 0 5
235 minsw4 t0, tf, t0 # 0 6
236 addq t3, t4, t3 # 1 3
240 maxsw4 t3, zero, t3 # 1 5
241 addq t6, t7, t6 # 2 3
245 minsw4 t3, tf, t3 # 1 6
248 maxsw4 t6, zero, t6 # 2 5
249 addq t9, ta, t9 # 3 3
251 minsw4 t6, tf, t6 # 2 6
254 maxsw4 t9, zero, t9 # 3 5
255 lda a0, 32(a0) # block += 16;
258 minsw4 t9, tf, t9 # 3 6
264 addq te, a2, a1 # pixels += line_size
270 .end add_pixels_clamped_mvi_asm