diff options
Diffstat (limited to 'erts/emulator/beam/erl_bits_f16.h')
| -rw-r--r-- | erts/emulator/beam/erl_bits_f16.h | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/erts/emulator/beam/erl_bits_f16.h b/erts/emulator/beam/erl_bits_f16.h index 653239d64d..cda52ab719 100644 --- a/erts/emulator/beam/erl_bits_f16.h +++ b/erts/emulator/beam/erl_bits_f16.h @@ -33,7 +33,7 @@ * 1. uint16_t and uint32_t have been rewritten to Uint16 and Uint32 * 2. Mixed declarations were moved to the top to avoid warnings * 3. inline was rewritten as ERTS_INLINE - * 4. UINT16_C(x) and UINT32_C(x) were rewriten to xU as we don't support 16bits platform + * 4. UINT16_C(x) and UINT32_C(x) were rewritten to xU as we don't support 16bits platform */ static ERTS_INLINE float fp32_from_bits(Uint32 w) { @@ -180,7 +180,7 @@ static ERTS_INLINE float fp16_ieee_to_fp32_value(Uint16 h) { * A normalized single-precision floating-point number is represented as: * FP32 = (1 + mantissa * 2**(-23)) * 2**(exponent - 127) * Therefore, when the biased exponent is 126, a unit change in the mantissa of the input denormalized half-precision - * number causes a change of the constructud single-precision number by 2**(-24), i.e. the same ammount. + * number causes a change of the constructud single-precision number by 2**(-24), i.e. the same amount. * * The last step is to adjust the bias of the constructed single-precision number. When the input half-precision number * is zero, the constructed single-precision number has the value of |