[trunk] Speed-up opj_int_fix_mul by removing unneeded operation
Update issue 220
This commit is contained in:
parent
e05d2901ec
commit
d0ce2ced53
@ -166,7 +166,7 @@ Multiply two fixed-precision rational numbers.
|
||||
*/
|
||||
static INLINE OPJ_INT32 opj_int_fix_mul(OPJ_INT32 a, OPJ_INT32 b) {
|
||||
OPJ_INT64 temp = (OPJ_INT64) a * (OPJ_INT64) b ;
|
||||
temp += temp & 4096;
|
||||
temp += 4096;
|
||||
return (OPJ_INT32) (temp >> 13) ;
|
||||
}
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user