Intel ITANIUM ARCHITECTURE - SOFTWARE DEVELOPERS MANUAL VOLUME 1 REV 2.3 Manual page 1182

Table 3-1. Pseudo-code Functions (Continued)
Function
fp_is_nan_or_inf(freg)
fp_is_natval(freg)
fp_is_normal(freg)
fp_is_pos_inf(freg)
fp_is_qnan(freg)
fp_is_snan(freg)
fp_is_unorm(freg)
fp_is_unsupported(freg)
fp_less_than(fr1, fr2)
fp_lesser_or_equal(fr1, fr2)
fp_mem_to_fr_format(mem, size)
fp_normalize(fr1)
fp_raise_fault(tmp_fp_env)
fp_raise_traps(tmp_fp_env)
fp_reg_bank_conflict(f1, f2)
fp_reg_disabled(f1, f2, f3, f4)
fp_reg_read(freg)
fp_unordered(fr1, fr2)
fp_update_fpsr(sf, tmp_fp_env)
fp_update_psr(dest_freg)
fpcmp_exception_fault_check(f2, f3, frel,
sf, *tmp_fp_env)
fpcvt_exception_fault_check(f2,
signed_form, trunc_form, sf,
*tmp_fp_env)
fpma_exception_fault_check(f2, f3, f4, sf,
*tmp_fp_env)
fpminmax_exception_fault_check(f2, f3,
sf, *tmp_fp_env)
fpms_fpnma_exception_fault_check(f2,
f3, f4, sf, *tmp_fp_env)
fprcpa_exception_fault_check(f2, f3, sf,
*tmp_fp_env, *limits_check)
fprsqrta_exception_fault_check(f3, sf,
*tmp_fp_env, *limits_check)
frcpa_exception_fault_check(f2, f3, sf,
*tmp_fp_env)
frsqrta_exception_fault_check(f3, sf,
*tmp_fp_env)
ignored_field_mask(regclass, reg, value)
Volume 3: Pseudo-Code Functions
Returns true if the floating-point exception_fault_check functions returned a IEEE
fault disabled default result or a propagated NaN.
Returns true when floating register contains a NaTVal
Returns true when floating register contains a normal number.
Returns true when floating register contains a positive infinity.
Returns true when floating register contains a quiet NaN.
Returns true when floating register contains a signalling NaN.
Returns true when floating register contains an unnormalized
number.
Returns true when floating register contains an unsupported format.
IEEE standard less-than relationship test.
IEEE standard less-than or equal-to relationship test
Converts a floating-point value in memory format to floating-point register format.
Normalizes an unnormalized fp value. This function flushes to zero any unnormal
values which can not be represented in the register file
Checks the local instruction state for any faulting conditions which require an
interruption to be raised.
Checks the local instruction state for any trapping conditions which require an
interruption to be raised.
Returns true if the two specified FRs are in the same bank.
Check for possible disabled floating-point register faults.
Reads the FR and gives canonical double-extended denormals (and
pseudo-denormals) their true mathematical exponent. Other classes of operands are
unaltered.
IEEE standard unordered relationship
Copies a floating-point instruction's local state into the global FPSR.
Conditionally sets PSR.mfl or PSR.mfh based on dest_freg.
Checks for all floating-point faulting conditions for the fpcmp instruction.
Checks for all floating-point faulting conditions for the fpcvt.fx , fpcvt.fxu ,
fpcvt.fx.trunc , and fpcvt.fxu.trunc instructions. It propagates NaNs.
Checks for all floating-point faulting conditions for the fpma instruction. It propagates
NaNs and special IEEE results.
Checks for all floating-point faulting conditions for the fpmin , fpmax , fpamin and
fpamax instructions.
Checks for all floating-point faulting conditions for the fpms and fpnma instructions. It
propagates NaNs and special IEEE results.
Checks for all floating-point faulting conditions for the fprcpa instruction. It
propagates NaNs and special IEEE results. It also indicates operand limit violations.
Checks for all floating-point faulting conditions for the fprsqrta instruction. It
propagates NaNs and special IEEE results. It also indicates operand limit violations.
Checks for all floating-point faulting conditions for the frcpa instruction. It
propagates NaNs and special IEEE results.
Checks for all floating-point faulting conditions for the frsqrta instruction. It
propagates NaNs and special IEEE results
Boolean function that returns value with bits cleared to 0 corresponding to ignored
bits for the specified register and register type.
Operation
3:283