Lines Matching refs:extended
1428 # "corrected" in order to have the proper equivalent extended precision
1587 # the extended precision result is still in fp0. but, we need to save it
1722 # underflow can happen for extended precision. extended precision opclass
2382 # _mem_read() - read extended immediate operand from memory #
2403 # (1) FP Instructions using extended precision or packed immediate #
2545 bsr.l decbin # convert to extended
2966 # instruction is using an extended precision immediate operand. Therefore,
3293 # For the case of an extended precision opclass 3 instruction, #
5310 # extended denorms are handled through another entry point.
6147 # a0 = pointer to extended precision input #
6698 # a0 = pointer to extended precision input #
6971 # extended-precision numbers whose square over/underflow #
7013 # L2 := extended-precision(-log2/64 - L1).#
7037 # 2^(J/64) to roughly 85 bits; T is in extended precision #
7081 # precision; and B2 is double extended. #
9455 tst.b %d0 # is precision extended?
9459 # Precision is extended
9569 # a0 = pointer to double-extended source operand X #
9570 # a1 = pointer to double-extended destination operand Y #
10166 # Entry point for scale w/ extended denorm. The function does
10180 andi.b &0xc0,%d1 # extended precision?
10183 # result precision is extended.
10280 beq.b ovfl_work # prec is extended
11536 # a0 = pointer to extended precision source operand #
11537 # a1 = pointer to extended precision destination operand #
11656 # extended precision. if the original operation was extended, then we have this
11658 # multiply using extended precision and the correct rounding mode. the result
11695 # - if precision is extended, then we have the EXOP. simply bias the exponent
11697 # calculate a result rounded to extended precision.
11767 # extended precision. if the original operation was extended, then we have this
11769 # multiply using extended precision and the correct rounding mode. the result
12032 # a0 = pointer to extended precision source operand #
12041 # norms into extended, single, and double precision. #
12353 # a0 = pointer to extended precision source operand #
12354 # a1 = pointer to extended precision destination operand #
13126 # a0 = pointer to extended precision source operand #
13211 # a0 = pointer to extended precision source operand #
13425 # a0 = pointer to extended precision source operand #
13434 # norms into extended, single, and double precision. #
13435 # Simply clear sign for extended precision norm. Ext prec denorm #
13469 andi.b &0xc0,%d0 # is precision extended?
13473 # precision selected is extended. so...we can not get an underflow
13486 # for an extended precision DENORM, the UNFL exception bit is set
13490 andi.b &0xc0,%d0 # is precision extended?
13508 # the input is an extended DENORM and underflow is enabled in the FPCR.
13614 # Therefore, we must return the result rounded to extended precision.
13671 # now, round to extended(and don't alter the FPSR).
13742 # a0 = pointer to extended precision source operand #
13743 # a1 = pointer to extended precision destination operand #
14741 andi.b &0xc0,%d1 # is precision extended?
14806 andi.b &0xc0,%d1 # is precision extended?
14840 # if the precision is extended, this result could not have come from an
15194 andi.b &0xc0,%d1 # is precision extended?
15259 andi.b &0xc0,%d1 # is precision extended?
15293 # if the precision is extended, this result could not have come from an
15514 # a0 = pointer to extended precision source operand #
15696 # Therefore, we must return the result rounded to extended precision.
15753 # now, round to extended(and don't alter the FPSR).
15962 # scale_to_zero_src(): scale the exponent of extended precision #
15969 # FP_SCR0(a6) = extended precision operand to be scaled #
15972 # FP_SCR0(a6) = scaled extended precision operand #
16089 # scale_to_zero_dst(): scale the exponent of extended precision #
16096 # FP_SCR1(a6) = extended precision operand to be scaled #
16099 # FP_SCR1(a6) = scaled extended precision operand #
16150 # FP_SRC(a6) = pointer to extended precision src operand #
16151 # FP_DST(a6) = pointer to extended precision dst operand #