^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) /* SPDX-License-Identifier: GPL-2.0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) #ifndef _VIDEO_ATAFB_UTILS_H
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) #define _VIDEO_ATAFB_UTILS_H
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) /* ================================================================= */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) /* Utility Assembler Functions */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) /* ================================================================= */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) /* ====================================================================== */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) /* Those of a delicate disposition might like to skip the next couple of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) * pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) * These functions are drop in replacements for memmove and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) * memset(_, 0, _). However their five instances add at least a kilobyte
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) * to the object file. You have been warned.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) * Not a great fan of assembler for the sake of it, but I think
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) * that these routines are at least 10 times faster than their C
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) * equivalents for large blits, and that's important to the lowest level of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) * a graphics driver. Question is whether some scheme with the blitter
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) * would be faster. I suspect not for simple text system - not much
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) * asynchrony.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) * Code is very simple, just gruesome expansion. Basic strategy is to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) * increase data moved/cleared at each step to 16 bytes to reduce
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) * instruction per data move overhead. movem might be faster still
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) * For more than 15 bytes, we try to align the write direction on a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) * longword boundary to get maximum speed. This is even more gruesome.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) * Unaligned read/write used requires 68020+ - think this is a problem?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) * Sorry!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) /* ++roman: I've optimized Robert's original versions in some minor
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) * aspects, e.g. moveq instead of movel, let gcc choose the registers,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) * use movem in some places...
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) * For other modes than 1 plane, lots of more such assembler functions
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) * were needed (e.g. the ones using movep or expanding color values).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) /* ++andreas: more optimizations:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) subl #65536,d0 replaced by clrw d0; subql #1,d0 for dbcc
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) addal is faster than addaw
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) movep is rather expensive compared to ordinary move's
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) some functions rewritten in C for clarity, no speed loss */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) static inline void *fb_memclear_small(void *s, size_t count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) if (!count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) " lsr.l #1,%1 ; jcc 1f ; move.b %2,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) "1: lsr.l #1,%1 ; jcc 1f ; move.w %2,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) "1: lsr.l #1,%1 ; jcc 1f ; move.l %2,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) "1: lsr.l #1,%1 ; jcc 1f ; move.l %2,-(%0) ; move.l %2,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) "1:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) : "=a" (s), "=d" (count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) : "d" (0), "0" ((char *)s + count), "1" (count));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) " subq.l #1,%1\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) " jcs 3f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) " move.l %2,%%d4; move.l %2,%%d5; move.l %2,%%d6\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) "2: movem.l %2/%%d4/%%d5/%%d6,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) " dbra %1,2b\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) "3:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) : "=a" (s), "=d" (count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) : "d" (0), "0" (s), "1" (count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) : "d4", "d5", "d6"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) );
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) static inline void *fb_memclear(void *s, size_t count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) if (!count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) if (count < 16) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) " lsr.l #1,%1 ; jcc 1f ; clr.b (%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) "1: lsr.l #1,%1 ; jcc 1f ; clr.w (%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) "1: lsr.l #1,%1 ; jcc 1f ; clr.l (%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) "1: lsr.l #1,%1 ; jcc 1f ; clr.l (%0)+ ; clr.l (%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) "1:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) : "=a" (s), "=d" (count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) : "0" (s), "1" (count));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) long tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) " move.l %1,%2\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) " lsr.l #1,%2 ; jcc 1f ; clr.b (%0)+ ; subq.w #1,%1\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) " lsr.l #1,%2 ; jcs 2f\n" /* %0 increased=>bit 2 switched*/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) " clr.w (%0)+ ; subq.w #2,%1 ; jra 2f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) "1: lsr.l #1,%2 ; jcc 2f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) " clr.w (%0)+ ; subq.w #2,%1\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) "2: move.w %1,%2; lsr.l #2,%1 ; jeq 6f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) " lsr.l #1,%1 ; jcc 3f ; clr.l (%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) "3: lsr.l #1,%1 ; jcc 4f ; clr.l (%0)+ ; clr.l (%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) "4: subq.l #1,%1 ; jcs 6f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) "5: clr.l (%0)+; clr.l (%0)+ ; clr.l (%0)+ ; clr.l (%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) " dbra %1,5b ; clr.w %1; subq.l #1,%1; jcc 5b\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) "6: move.w %2,%1; btst #1,%1 ; jeq 7f ; clr.w (%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) "7: btst #0,%1 ; jeq 8f ; clr.b (%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) "8:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) : "=a" (s), "=d" (count), "=d" (tmp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) : "0" (s), "1" (count));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) static inline void *fb_memset255(void *s, size_t count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) if (!count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) " lsr.l #1,%1 ; jcc 1f ; move.b %2,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) "1: lsr.l #1,%1 ; jcc 1f ; move.w %2,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) "1: lsr.l #1,%1 ; jcc 1f ; move.l %2,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) "1: lsr.l #1,%1 ; jcc 1f ; move.l %2,-(%0) ; move.l %2,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) "1:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) : "=a" (s), "=d" (count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) : "d" (-1), "0" ((char *)s+count), "1" (count));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) " subq.l #1,%1 ; jcs 3f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) " move.l %2,%%d4; move.l %2,%%d5; move.l %2,%%d6\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) "2: movem.l %2/%%d4/%%d5/%%d6,-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) " dbra %1,2b\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) "3:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) : "=a" (s), "=d" (count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) : "d" (-1), "0" (s), "1" (count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) : "d4", "d5", "d6");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) static inline void *fb_memmove(void *d, const void *s, size_t count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) if (d < s) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) if (count < 16) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) " lsr.l #1,%2 ; jcc 1f ; move.b (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) "1: lsr.l #1,%2 ; jcc 1f ; move.w (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) "1: lsr.l #1,%2 ; jcc 1f ; move.l (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) "1: lsr.l #1,%2 ; jcc 1f ; move.l (%1)+,(%0)+ ; move.l (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) "1:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) : "=a" (d), "=a" (s), "=d" (count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) : "0" (d), "1" (s), "2" (count));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) long tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) " move.l %0,%3\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) " lsr.l #1,%3 ; jcc 1f ; move.b (%1)+,(%0)+ ; subqw #1,%2\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) " lsr.l #1,%3 ; jcs 2f\n" /* %0 increased=>bit 2 switched*/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) " move.w (%1)+,(%0)+ ; subqw #2,%2 ; jra 2f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) "1: lsr.l #1,%3 ; jcc 2f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) " move.w (%1)+,(%0)+ ; subqw #2,%2\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) "2: move.w %2,%-; lsr.l #2,%2 ; jeq 6f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) " lsr.l #1,%2 ; jcc 3f ; move.l (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) "3: lsr.l #1,%2 ; jcc 4f ; move.l (%1)+,(%0)+ ; move.l (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) "4: subq.l #1,%2 ; jcs 6f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) "5: move.l (%1)+,(%0)+; move.l (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) " move.l (%1)+,(%0)+; move.l (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) " dbra %2,5b ; clr.w %2; subq.l #1,%2; jcc 5b\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) "6: move.w %+,%2; btst #1,%2 ; jeq 7f ; move.w (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) "7: btst #0,%2 ; jeq 8f ; move.b (%1)+,(%0)+\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) "8:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) : "=a" (d), "=a" (s), "=d" (count), "=d" (tmp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) : "0" (d), "1" (s), "2" (count));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) if (count < 16) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) " lsr.l #1,%2 ; jcc 1f ; move.b -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) "1: lsr.l #1,%2 ; jcc 1f ; move.w -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) "1: lsr.l #1,%2 ; jcc 1f ; move.l -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) "1: lsr.l #1,%2 ; jcc 1f ; move.l -(%1),-(%0) ; move.l -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) "1:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) : "=a" (d), "=a" (s), "=d" (count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) : "0" ((char *) d + count), "1" ((char *) s + count), "2" (count));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) long tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) " move.l %0,%3\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) " lsr.l #1,%3 ; jcc 1f ; move.b -(%1),-(%0) ; subqw #1,%2\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) " lsr.l #1,%3 ; jcs 2f\n" /* %0 increased=>bit 2 switched*/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) " move.w -(%1),-(%0) ; subqw #2,%2 ; jra 2f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) "1: lsr.l #1,%3 ; jcc 2f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) " move.w -(%1),-(%0) ; subqw #2,%2\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) "2: move.w %2,%-; lsr.l #2,%2 ; jeq 6f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) " lsr.l #1,%2 ; jcc 3f ; move.l -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) "3: lsr.l #1,%2 ; jcc 4f ; move.l -(%1),-(%0) ; move.l -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) "4: subq.l #1,%2 ; jcs 6f\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) "5: move.l -(%1),-(%0); move.l -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) " move.l -(%1),-(%0); move.l -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) " dbra %2,5b ; clr.w %2; subq.l #1,%2; jcc 5b\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) "6: move.w %+,%2; btst #1,%2 ; jeq 7f ; move.w -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) "7: btst #0,%2 ; jeq 8f ; move.b -(%1),-(%0)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) "8:"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) : "=a" (d), "=a" (s), "=d" (count), "=d" (tmp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) : "0" ((char *) d + count), "1" ((char *) s + count), "2" (count));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) /* ++andreas: Simple and fast version of memmove, assumes size is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) divisible by 16, suitable for moving the whole screen bitplane */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) static inline void fast_memmove(char *dst, const char *src, size_t size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) if (!size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) if (dst < src)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) "1: movem.l (%0)+,%%d0/%%d1/%%a0/%%a1\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) " movem.l %%d0/%%d1/%%a0/%%a1,%1@\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) " addq.l #8,%1; addq.l #8,%1\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) " dbra %2,1b\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) " clr.w %2; subq.l #1,%2\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) " jcc 1b"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) : "=a" (src), "=a" (dst), "=d" (size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) : "0" (src), "1" (dst), "2" (size / 16 - 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) : "d0", "d1", "a0", "a1", "memory");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) asm volatile ("\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) "1: subq.l #8,%0; subq.l #8,%0\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) " movem.l %0@,%%d0/%%d1/%%a0/%%a1\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) " movem.l %%d0/%%d1/%%a0/%%a1,-(%1)\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) " dbra %2,1b\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) " clr.w %2; subq.l #1,%2\n"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) " jcc 1b"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) : "=a" (src), "=a" (dst), "=d" (size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) : "0" (src + size), "1" (dst + size), "2" (size / 16 - 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) : "d0", "d1", "a0", "a1", "memory");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) #ifdef BPL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) * This expands a up to 8 bit color into two longs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) * for movel operations.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) static const u32 four2long[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) 0x00000000, 0x000000ff, 0x0000ff00, 0x0000ffff,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) 0x00ff0000, 0x00ff00ff, 0x00ffff00, 0x00ffffff,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) 0xff000000, 0xff0000ff, 0xff00ff00, 0xff00ffff,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) 0xffff0000, 0xffff00ff, 0xffffff00, 0xffffffff,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) static inline void expand8_col2mask(u8 c, u32 m[])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) m[0] = four2long[c & 15];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) #if BPL > 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) m[1] = four2long[c >> 4];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) static inline void expand8_2col2mask(u8 fg, u8 bg, u32 fgm[], u32 bgm[])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) fgm[0] = four2long[fg & 15] ^ (bgm[0] = four2long[bg & 15]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) #if BPL > 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) fgm[1] = four2long[fg >> 4] ^ (bgm[1] = four2long[bg >> 4]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) * set an 8bit value to a color
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) static inline void fill8_col(u8 *dst, u32 m[])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) u32 tmp = m[0];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) dst[0] = tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) dst[2] = (tmp >>= 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) #if BPL > 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) dst[4] = (tmp >>= 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) dst[6] = tmp >> 8;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) #if BPL > 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) tmp = m[1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) dst[8] = tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) dst[10] = (tmp >>= 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) dst[12] = (tmp >>= 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) dst[14] = tmp >> 8;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) * set an 8bit value according to foreground/background color
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) static inline void fill8_2col(u8 *dst, u8 fg, u8 bg, u32 mask)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) u32 fgm[2], bgm[2], tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) expand8_2col2mask(fg, bg, fgm, bgm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) mask |= mask << 8;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) #if BPL > 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) mask |= mask << 16;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) tmp = (mask & fgm[0]) ^ bgm[0];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) dst[0] = tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) dst[2] = (tmp >>= 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) #if BPL > 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) dst[4] = (tmp >>= 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) dst[6] = tmp >> 8;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) #if BPL > 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) tmp = (mask & fgm[1]) ^ bgm[1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) dst[8] = tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) dst[10] = (tmp >>= 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) dst[12] = (tmp >>= 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) dst[14] = tmp >> 8;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) static const u32 two2word[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) 0x00000000, 0xffff0000, 0x0000ffff, 0xffffffff
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) static inline void expand16_col2mask(u8 c, u32 m[])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) m[0] = two2word[c & 3];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) #if BPL > 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) m[1] = two2word[(c >> 2) & 3];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) #if BPL > 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) m[2] = two2word[(c >> 4) & 3];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) m[3] = two2word[c >> 6];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) static inline void expand16_2col2mask(u8 fg, u8 bg, u32 fgm[], u32 bgm[])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) bgm[0] = two2word[bg & 3];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) fgm[0] = two2word[fg & 3] ^ bgm[0];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) #if BPL > 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) bgm[1] = two2word[(bg >> 2) & 3];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) fgm[1] = two2word[(fg >> 2) & 3] ^ bgm[1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) #if BPL > 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) bgm[2] = two2word[(bg >> 4) & 3];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) fgm[2] = two2word[(fg >> 4) & 3] ^ bgm[2];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) bgm[3] = two2word[bg >> 6];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) fgm[3] = two2word[fg >> 6] ^ bgm[3];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) static inline u32 *fill16_col(u32 *dst, int rows, u32 m[])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) while (rows) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) *dst++ = m[0];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) #if BPL > 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) *dst++ = m[1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) #if BPL > 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) *dst++ = m[2];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) *dst++ = m[3];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) rows--;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) return dst;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) static inline void memmove32_col(void *dst, void *src, u32 mask, u32 h, u32 bytes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) u32 *s, *d, v;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) s = src;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) d = dst;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) v = (*s++ & mask) | (*d & ~mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) *d++ = v;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) #if BPL > 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) v = (*s++ & mask) | (*d & ~mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) *d++ = v;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) #if BPL > 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) v = (*s++ & mask) | (*d & ~mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) *d++ = v;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) v = (*s++ & mask) | (*d & ~mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) *d++ = v;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) d = (u32 *)((u8 *)d + bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) s = (u32 *)((u8 *)s + bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) } while (--h);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) #endif /* _VIDEO_ATAFB_UTILS_H */
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags