### 23.12.5. FDIV

The thresholds for divide are simple and based only on the difference of the exponents of the dividend and the divisor. It is not possible in a divide operation for the significand to overflow and cause an increment of the exponent. However, it is possible for the significand to require a single bit left shift and the exponent to be decremented for normalization. To reduce logic complexity, the overflow ranges are the same as those of the LSA operations in FADD and FSUB. The underflow ranges include the minimum normal exponent, `0x01` for single-precision and `0x001` for double-precision. Table 23.10 shows the FDIV bounce thresholds. The exponent values shown in Table 23.10 are in biased format.

Table 23.10. FDIV bounce thresholds

Initial quotient exponent valueFloat valueCondition in full-compliance mode
DP[1]SP[2]SPDP
>`0x7FF`-DP overflow-Bounce
`0x7FF`-DP NaN or infinity-Bounce
`0x7FE`-DP maximum normal-Bounce
`0x7FD`-DP normal-Bounce
`0x7FC`-DP normal-Normal
>`0x47F`>`0xFF`SP overflowBounceNormal
`0x47F``0xFF`SP NaN or infinityBounceNormal
`0x47E``0xFE`SP maximum normalBounceNormal
`0x47D``0xFD`SP normalBounceNormal
`0x47C``0xFC`SP normalNormalNormal
`0x3FF``0x7F`e = 0 bias value NormalNormal
`0x382``0x02`SP normalNormalNormal
`0x381``0x01`SP normalBounceNormal
`0x380``0x00`SP subnormalBounceNormal
<`0x380`<`0x00`SP underflowBounceNormal
`0x002`-DP normal-Normal
`0x001`-DP normal-Bounce
`0x000`-DP subnormal-Bounce
<`0x000`-DP underflow-Bounce

[1] DP = double-precision.

[2] SP = single-precision.