1 /* { dg-require-effective-target vect_int } */
10 unsigned char X
[N
] __attribute__ ((__aligned__(__BIGGEST_ALIGNMENT__
))) = {0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63};
11 unsigned char Y
[N
] __attribute__ ((__aligned__(__BIGGEST_ALIGNMENT__
))) = {64,63,62,61,60,59,58,57,56,55,54,53,52,51,50,49,48,47,46,45,44,43,42,41,40,39,38,37,36,35,34,33,32,31,30,29,28,27,26,25,24,23,22,21,20,19,18,17,16,15,14,13,12,11,10,9,8,7,6,5,4,3,2,1};
13 /* char->short->short dot product.
14 Detected as a dot-product pattern.
15 Should be vectorized on targets that support dot-product for unsigned chars,
16 but currently this test cannot be vectorized as a dot-product on targets
17 that support char->short->int dot-product.
18 Alternatively, this test can be vectorized using vect_widen_mult_qi (or
19 vect_unpack and non-widening multplication: vect_unpack && vect_short_mult).
21 __attribute__ ((noinline
)) unsigned short
24 unsigned short result
= 0;
26 for (i
=0; i
<len
; i
++) {
27 result
+= (unsigned short)(X
[i
] * Y
[i
]);
46 /* { dg-final { scan-tree-dump-times "vect_recog_dot_prod_pattern: detected" 1 "vect" } } */
48 /* When the vectorizer is enhanced to vectorize accumulation into short for
49 targets that support accumulation into int (powerpc, ia64) we'd have:
50 dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { target vect_udot_qi || vect_widen_mult_qi_to_hi } }
52 /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" {target { vect_widen_mult_qi_to_hi || vect_unpack } } } } */
54 /* { dg-final { cleanup-tree-dump "vect" } } */