Intel Itanium 9150M, Manual

Page: 88 / 420

Volume 3: Instruction Reference 3:79
fmax
fmax — Floating-point Maximum
Format: (
qp
) fmax.
sf f
1
=
f
2
,
f
3
F8
Description:
The operand with the larger value is placed in FR
f
1
. If FR
f
2
equals FR
f
3
, FR
f
1
gets FR
f
3
.
If either FR
f
2
or FR
f
3
is a NaN, FR
f
1
gets FR
f
3
.
If either FR
f
2
or FR
f
3
is a NaTVal, FR
f
1
is set to NaTVal instead of the computed result.
This operation does not propagate NaNs the same way as other arithmetic
floating-point instructions. The Invalid Operation is signaled in the same manner as the
fcmp.lt
operation.
The mnemonic values for
sf
are given in Table 2-23 on page 3:56 .
Operation: if (PR[ qp ]) {
fp_check_target_register( f
1
);
if (tmp_isrcode = fp_reg_disabled( f
1
, f
2
, f
3
, 0))
disabled_fp_register_fault(tmp_isrcode, 0);
if (fp_is_natval(FR[ f
2
]) || fp_is_natval(FR[ f
3
])) {
FR[ f
1
] = NATVAL;
} else {
fminmax_exception_fault_check( f
2
, f
3
, sf, &tmp_fp_env);
if (fp_raise_fault(tmp_fp_env))
fp_exception_fault(fp_decode_fault(tmp_fp_env));
tmp_bool_res = fp_less_than(fp_reg_read(FR[ f
3
]),
fp_reg_read(FR[ f
2
]));
FR[ f
1
] = (tmp_bool_res ? FR[ f
2
] : FR[ f
3
]);
fp_update_fpsr( sf , tmp_fp_env);
}
fp_update_psr(
f
1
);
}
FP Exceptions:
Invalid Operation (V)
Denormal/Unnormal Operand (D)
Software Assist (SWA) fault
Interruptions:
Illegal Operation fault Floating-point Exception fault
Disabled Floating-point Register fault