.type memcpy, @function
.ent memcpy
+#ifdef __MICROBLAZEEL__
+# define BSLLI bsrli
+# define BSRLI bslli
+#else
+# define BSLLI bslli
+# define BSRLI bsrli
+#endif
+
memcpy:
fast_memcpy_ascending:
/* move d to return register as value of function */
beqi r9, a_block_u2 /* t1 was 2 => 2 byte offset */
a_block_u3:
- bslli r11, r11, 24 /* h = h << 24 */
+ BSLLI r11, r11, 24 /* h = h << 24 */
a_bu3_loop:
lwi r12, r8, 4 /* v = *(as + 4) */
- bsrli r9, r12, 8 /* t1 = v >> 8 */
+ BSRLI r9, r12, 8 /* t1 = v >> 8 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 0 /* *(d + 0) = t1 */
- bslli r11, r12, 24 /* h = v << 24 */
+ BSLLI r11, r12, 24 /* h = v << 24 */
lwi r12, r8, 8 /* v = *(as + 8) */
- bsrli r9, r12, 8 /* t1 = v >> 8 */
+ BSRLI r9, r12, 8 /* t1 = v >> 8 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 4 /* *(d + 4) = t1 */
- bslli r11, r12, 24 /* h = v << 24 */
+ BSLLI r11, r12, 24 /* h = v << 24 */
lwi r12, r8, 12 /* v = *(as + 12) */
- bsrli r9, r12, 8 /* t1 = v >> 8 */
+ BSRLI r9, r12, 8 /* t1 = v >> 8 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 8 /* *(d + 8) = t1 */
- bslli r11, r12, 24 /* h = v << 24 */
+ BSLLI r11, r12, 24 /* h = v << 24 */
lwi r12, r8, 16 /* v = *(as + 16) */
- bsrli r9, r12, 8 /* t1 = v >> 8 */
+ BSRLI r9, r12, 8 /* t1 = v >> 8 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 12 /* *(d + 12) = t1 */
- bslli r11, r12, 24 /* h = v << 24 */
+ BSLLI r11, r12, 24 /* h = v << 24 */
lwi r12, r8, 20 /* v = *(as + 20) */
- bsrli r9, r12, 8 /* t1 = v >> 8 */
+ BSRLI r9, r12, 8 /* t1 = v >> 8 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 16 /* *(d + 16) = t1 */
- bslli r11, r12, 24 /* h = v << 24 */
+ BSLLI r11, r12, 24 /* h = v << 24 */
lwi r12, r8, 24 /* v = *(as + 24) */
- bsrli r9, r12, 8 /* t1 = v >> 8 */
+ BSRLI r9, r12, 8 /* t1 = v >> 8 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 20 /* *(d + 20) = t1 */
- bslli r11, r12, 24 /* h = v << 24 */
+ BSLLI r11, r12, 24 /* h = v << 24 */
lwi r12, r8, 28 /* v = *(as + 28) */
- bsrli r9, r12, 8 /* t1 = v >> 8 */
+ BSRLI r9, r12, 8 /* t1 = v >> 8 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 24 /* *(d + 24) = t1 */
- bslli r11, r12, 24 /* h = v << 24 */
+ BSLLI r11, r12, 24 /* h = v << 24 */
lwi r12, r8, 32 /* v = *(as + 32) */
- bsrli r9, r12, 8 /* t1 = v >> 8 */
+ BSRLI r9, r12, 8 /* t1 = v >> 8 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 28 /* *(d + 28) = t1 */
- bslli r11, r12, 24 /* h = v << 24 */
+ BSLLI r11, r12, 24 /* h = v << 24 */
addi r8, r8, 32 /* as = as + 32 */
addi r4, r4, -32 /* n = n - 32 */
bneid r4, a_bu3_loop /* while (n) loop */
bri a_block_done
a_block_u1:
- bslli r11, r11, 8 /* h = h << 8 */
+ BSLLI r11, r11, 8 /* h = h << 8 */
a_bu1_loop:
lwi r12, r8, 4 /* v = *(as + 4) */
- bsrli r9, r12, 24 /* t1 = v >> 24 */
+ BSRLI r9, r12, 24 /* t1 = v >> 24 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 0 /* *(d + 0) = t1 */
- bslli r11, r12, 8 /* h = v << 8 */
+ BSLLI r11, r12, 8 /* h = v << 8 */
lwi r12, r8, 8 /* v = *(as + 8) */
- bsrli r9, r12, 24 /* t1 = v >> 24 */
+ BSRLI r9, r12, 24 /* t1 = v >> 24 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 4 /* *(d + 4) = t1 */
- bslli r11, r12, 8 /* h = v << 8 */
+ BSLLI r11, r12, 8 /* h = v << 8 */
lwi r12, r8, 12 /* v = *(as + 12) */
- bsrli r9, r12, 24 /* t1 = v >> 24 */
+ BSRLI r9, r12, 24 /* t1 = v >> 24 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 8 /* *(d + 8) = t1 */
- bslli r11, r12, 8 /* h = v << 8 */
+ BSLLI r11, r12, 8 /* h = v << 8 */
lwi r12, r8, 16 /* v = *(as + 16) */
- bsrli r9, r12, 24 /* t1 = v >> 24 */
+ BSRLI r9, r12, 24 /* t1 = v >> 24 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 12 /* *(d + 12) = t1 */
- bslli r11, r12, 8 /* h = v << 8 */
+ BSLLI r11, r12, 8 /* h = v << 8 */
lwi r12, r8, 20 /* v = *(as + 20) */
- bsrli r9, r12, 24 /* t1 = v >> 24 */
+ BSRLI r9, r12, 24 /* t1 = v >> 24 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 16 /* *(d + 16) = t1 */
- bslli r11, r12, 8 /* h = v << 8 */
+ BSLLI r11, r12, 8 /* h = v << 8 */
lwi r12, r8, 24 /* v = *(as + 24) */
- bsrli r9, r12, 24 /* t1 = v >> 24 */
+ BSRLI r9, r12, 24 /* t1 = v >> 24 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 20 /* *(d + 20) = t1 */
- bslli r11, r12, 8 /* h = v << 8 */
+ BSLLI r11, r12, 8 /* h = v << 8 */
lwi r12, r8, 28 /* v = *(as + 28) */
- bsrli r9, r12, 24 /* t1 = v >> 24 */
+ BSRLI r9, r12, 24 /* t1 = v >> 24 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 24 /* *(d + 24) = t1 */
- bslli r11, r12, 8 /* h = v << 8 */
+ BSLLI r11, r12, 8 /* h = v << 8 */
lwi r12, r8, 32 /* v = *(as + 32) */
- bsrli r9, r12, 24 /* t1 = v >> 24 */
+ BSRLI r9, r12, 24 /* t1 = v >> 24 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 28 /* *(d + 28) = t1 */
- bslli r11, r12, 8 /* h = v << 8 */
+ BSLLI r11, r12, 8 /* h = v << 8 */
addi r8, r8, 32 /* as = as + 32 */
addi r4, r4, -32 /* n = n - 32 */
bneid r4, a_bu1_loop /* while (n) loop */
bri a_block_done
a_block_u2:
- bslli r11, r11, 16 /* h = h << 16 */
+ BSLLI r11, r11, 16 /* h = h << 16 */
a_bu2_loop:
lwi r12, r8, 4 /* v = *(as + 4) */
- bsrli r9, r12, 16 /* t1 = v >> 16 */
+ BSRLI r9, r12, 16 /* t1 = v >> 16 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 0 /* *(d + 0) = t1 */
- bslli r11, r12, 16 /* h = v << 16 */
+ BSLLI r11, r12, 16 /* h = v << 16 */
lwi r12, r8, 8 /* v = *(as + 8) */
- bsrli r9, r12, 16 /* t1 = v >> 16 */
+ BSRLI r9, r12, 16 /* t1 = v >> 16 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 4 /* *(d + 4) = t1 */
- bslli r11, r12, 16 /* h = v << 16 */
+ BSLLI r11, r12, 16 /* h = v << 16 */
lwi r12, r8, 12 /* v = *(as + 12) */
- bsrli r9, r12, 16 /* t1 = v >> 16 */
+ BSRLI r9, r12, 16 /* t1 = v >> 16 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 8 /* *(d + 8) = t1 */
- bslli r11, r12, 16 /* h = v << 16 */
+ BSLLI r11, r12, 16 /* h = v << 16 */
lwi r12, r8, 16 /* v = *(as + 16) */
- bsrli r9, r12, 16 /* t1 = v >> 16 */
+ BSRLI r9, r12, 16 /* t1 = v >> 16 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 12 /* *(d + 12) = t1 */
- bslli r11, r12, 16 /* h = v << 16 */
+ BSLLI r11, r12, 16 /* h = v << 16 */
lwi r12, r8, 20 /* v = *(as + 20) */
- bsrli r9, r12, 16 /* t1 = v >> 16 */
+ BSRLI r9, r12, 16 /* t1 = v >> 16 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 16 /* *(d + 16) = t1 */
- bslli r11, r12, 16 /* h = v << 16 */
+ BSLLI r11, r12, 16 /* h = v << 16 */
lwi r12, r8, 24 /* v = *(as + 24) */
- bsrli r9, r12, 16 /* t1 = v >> 16 */
+ BSRLI r9, r12, 16 /* t1 = v >> 16 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 20 /* *(d + 20) = t1 */
- bslli r11, r12, 16 /* h = v << 16 */
+ BSLLI r11, r12, 16 /* h = v << 16 */
lwi r12, r8, 28 /* v = *(as + 28) */
- bsrli r9, r12, 16 /* t1 = v >> 16 */
+ BSRLI r9, r12, 16 /* t1 = v >> 16 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 24 /* *(d + 24) = t1 */
- bslli r11, r12, 16 /* h = v << 16 */
+ BSLLI r11, r12, 16 /* h = v << 16 */
lwi r12, r8, 32 /* v = *(as + 32) */
- bsrli r9, r12, 16 /* t1 = v >> 16 */
+ BSRLI r9, r12, 16 /* t1 = v >> 16 */
or r9, r11, r9 /* t1 = h | t1 */
swi r9, r5, 28 /* *(d + 28) = t1 */
- bslli r11, r12, 16 /* h = v << 16 */
+ BSLLI r11, r12, 16 /* h = v << 16 */
addi r8, r8, 32 /* as = as + 32 */
addi r4, r4, -32 /* n = n - 32 */
bneid r4, a_bu2_loop /* while (n) loop */
beqi r9, a_word_u2 /* t1 was 2 => 2 byte offset */
a_word_u3:
- bslli r11, r11, 24 /* h = h << 24 */
+ BSLLI r11, r11, 24 /* h = h << 24 */
a_wu3_loop:
lw r12, r8, r10 /* v = *(as + offset) */
- bsrli r9, r12, 8 /* t1 = v >> 8 */
+ BSRLI r9, r12, 8 /* t1 = v >> 8 */
or r9, r11, r9 /* t1 = h | t1 */
sw r9, r5, r10 /* *(d + offset) = t1 */
- bslli r11, r12, 24 /* h = v << 24 */
+ BSLLI r11, r12, 24 /* h = v << 24 */
addi r4, r4,-4 /* n = n - 4 */
bneid r4, a_wu3_loop /* while (n) loop */
addi r10, r10, 4 /* offset = ofset + 4 (IN DELAY SLOT) */
bri a_word_done
a_word_u1:
- bslli r11, r11, 8 /* h = h << 8 */
+ BSLLI r11, r11, 8 /* h = h << 8 */
a_wu1_loop:
lw r12, r8, r10 /* v = *(as + offset) */
- bsrli r9, r12, 24 /* t1 = v >> 24 */
+ BSRLI r9, r12, 24 /* t1 = v >> 24 */
or r9, r11, r9 /* t1 = h | t1 */
sw r9, r5, r10 /* *(d + offset) = t1 */
- bslli r11, r12, 8 /* h = v << 8 */
+ BSLLI r11, r12, 8 /* h = v << 8 */
addi r4, r4,-4 /* n = n - 4 */
bneid r4, a_wu1_loop /* while (n) loop */
addi r10, r10, 4 /* offset = ofset + 4 (IN DELAY SLOT) */
bri a_word_done
a_word_u2:
- bslli r11, r11, 16 /* h = h << 16 */
+ BSLLI r11, r11, 16 /* h = h << 16 */
a_wu2_loop:
lw r12, r8, r10 /* v = *(as + offset) */
- bsrli r9, r12, 16 /* t1 = v >> 16 */
+ BSRLI r9, r12, 16 /* t1 = v >> 16 */
or r9, r11, r9 /* t1 = h | t1 */
sw r9, r5, r10 /* *(d + offset) = t1 */
- bslli r11, r12, 16 /* h = v << 16 */
+ BSLLI r11, r12, 16 /* h = v << 16 */
addi r4, r4,-4 /* n = n - 4 */
bneid r4, a_wu2_loop /* while (n) loop */
addi r10, r10, 4 /* offset = ofset + 4 (IN DELAY SLOT) */