| Message ID | 20180124131315.30567-12-alex.bennee@linaro.org |
|---|---|
| State | Superseded |
| Headers | show |
| Series | re-factor softfloat and add fp16 functions | expand |
On 01/24/2018 10:13 AM, Alex Bennée wrote: > These structures pave the way for generic softfloat helper routines > that will operate on fully decomposed numbers. I have to say this patch in particular is very elegant (seeing how it simplify the later refactors). I suppose you had a long brainstorming before... Total-brain-hours-spent: 141 > Signed-off-by: Alex Bennée <alex.bennee@linaro.org> > Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> > > --- > v3 > - comment box style > - CamelCase structs > - hide DECOMPOSED_BINARY_POINT - frac in macro > - more comments > - add exp_size, frac_size to FloatFmt > - compute exp_bias and exp_max from FLOAT_PARAMS > - remove include bitops (in next patch) > --- > fpu/softfloat.c | 86 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++- > 1 file changed, 85 insertions(+), 1 deletion(-) > > diff --git a/fpu/softfloat.c b/fpu/softfloat.c > index 297e48f5c9..568d555595 100644 > --- a/fpu/softfloat.c > +++ b/fpu/softfloat.c > @@ -83,7 +83,6 @@ this code that are retained. > * target-dependent and needs the TARGET_* macros. > */ > #include "qemu/osdep.h" > - > #include "fpu/softfloat.h" > > /* We only need stdlib for abort() */ > @@ -186,6 +185,91 @@ static inline flag extractFloat64Sign(float64 a) > return float64_val(a) >> 63; > } > > +/* > + * Classify a floating point number. Everything above float_class_qnan > + * is a NaN so cls >= float_class_qnan is any NaN. > + */ > + > +typedef enum __attribute__ ((__packed__)) { > + float_class_unclassified, > + float_class_zero, > + float_class_normal, > + float_class_inf, > + float_class_qnan, /* all NaNs from here */ > + float_class_snan, > + float_class_dnan, > + float_class_msnan, /* maybe silenced */ > +} FloatClass; > + > +/* > + * Structure holding all of the decomposed parts of a float. The > + * exponent is unbiased and the fraction is normalized. All > + * calculations are done with a 64 bit fraction and then rounded as > + * appropriate for the final format. > + * > + * Thanks to the packed FloatClass a decent compiler should be able to > + * fit the whole structure into registers and avoid using the stack > + * for parameter passing. > + */ > + > +typedef struct { > + uint64_t frac; > + int32_t exp; > + FloatClass cls; > + bool sign; > +} FloatParts; > + > +#define DECOMPOSED_BINARY_POINT (64 - 2) > +#define DECOMPOSED_IMPLICIT_BIT (1ull << DECOMPOSED_BINARY_POINT) > +#define DECOMPOSED_OVERFLOW_BIT (DECOMPOSED_IMPLICIT_BIT << 1) > + > +/* Structure holding all of the relevant parameters for a format. > + * exp_size: the size of the exponent field > + * exp_bias: the offset applied to the exponent field > + * exp_max: the maximum normalised exponent > + * frac_size: the size of the fraction field > + * frac_shift: shift to normalise the fraction with DECOMPOSED_BINARY_POINT > + * The following are computed based the size of fraction > + * frac_lsb: least significant bit of fraction > + * fram_lsbm1: the bit bellow the least significant bit (for rounding) > + * round_mask/roundeven_mask: masks used for rounding > + */ > +typedef struct { > + int exp_size; > + int exp_bias; > + int exp_max; > + int frac_size; > + int frac_shift; > + uint64_t frac_lsb; > + uint64_t frac_lsbm1; > + uint64_t round_mask; > + uint64_t roundeven_mask; > +} FloatFmt; > + > +/* Expand fields based on the size of exponent and fraction */ > +#define FLOAT_PARAMS(E, F) \ > + .exp_size = E, \ > + .exp_bias = ((1 << E) - 1) >> 1, \ > + .exp_max = (1 << E) - 1, \ > + .frac_size = F, \ > + .frac_shift = DECOMPOSED_BINARY_POINT - F, \ > + .frac_lsb = 1ull << (DECOMPOSED_BINARY_POINT - F), \ > + .frac_lsbm1 = 1ull << ((DECOMPOSED_BINARY_POINT - F) - 1), \ > + .round_mask = (1ull << (DECOMPOSED_BINARY_POINT - F)) - 1, \ > + .roundeven_mask = (2ull << (DECOMPOSED_BINARY_POINT - F)) - 1 > + > +static const FloatFmt float16_params = { > + FLOAT_PARAMS(5, 10) > +}; > + > +static const FloatFmt float32_params = { > + FLOAT_PARAMS(8, 23) > +}; > + > +static const FloatFmt float64_params = { > + FLOAT_PARAMS(11, 52) > +}; > + > /*---------------------------------------------------------------------------- > | Takes a 64-bit fixed-point value `absZ' with binary point between bits 6 > | and 7, and returns the properly rounded 32-bit integer corresponding to the >
diff --git a/fpu/softfloat.c b/fpu/softfloat.c index 297e48f5c9..568d555595 100644 --- a/fpu/softfloat.c +++ b/fpu/softfloat.c @@ -83,7 +83,6 @@ this code that are retained. * target-dependent and needs the TARGET_* macros. */ #include "qemu/osdep.h" - #include "fpu/softfloat.h" /* We only need stdlib for abort() */ @@ -186,6 +185,91 @@ static inline flag extractFloat64Sign(float64 a) return float64_val(a) >> 63; } +/* + * Classify a floating point number. Everything above float_class_qnan + * is a NaN so cls >= float_class_qnan is any NaN. + */ + +typedef enum __attribute__ ((__packed__)) { + float_class_unclassified, + float_class_zero, + float_class_normal, + float_class_inf, + float_class_qnan, /* all NaNs from here */ + float_class_snan, + float_class_dnan, + float_class_msnan, /* maybe silenced */ +} FloatClass; + +/* + * Structure holding all of the decomposed parts of a float. The + * exponent is unbiased and the fraction is normalized. All + * calculations are done with a 64 bit fraction and then rounded as + * appropriate for the final format. + * + * Thanks to the packed FloatClass a decent compiler should be able to + * fit the whole structure into registers and avoid using the stack + * for parameter passing. + */ + +typedef struct { + uint64_t frac; + int32_t exp; + FloatClass cls; + bool sign; +} FloatParts; + +#define DECOMPOSED_BINARY_POINT (64 - 2) +#define DECOMPOSED_IMPLICIT_BIT (1ull << DECOMPOSED_BINARY_POINT) +#define DECOMPOSED_OVERFLOW_BIT (DECOMPOSED_IMPLICIT_BIT << 1) + +/* Structure holding all of the relevant parameters for a format. + * exp_size: the size of the exponent field + * exp_bias: the offset applied to the exponent field + * exp_max: the maximum normalised exponent + * frac_size: the size of the fraction field + * frac_shift: shift to normalise the fraction with DECOMPOSED_BINARY_POINT + * The following are computed based the size of fraction + * frac_lsb: least significant bit of fraction + * fram_lsbm1: the bit bellow the least significant bit (for rounding) + * round_mask/roundeven_mask: masks used for rounding + */ +typedef struct { + int exp_size; + int exp_bias; + int exp_max; + int frac_size; + int frac_shift; + uint64_t frac_lsb; + uint64_t frac_lsbm1; + uint64_t round_mask; + uint64_t roundeven_mask; +} FloatFmt; + +/* Expand fields based on the size of exponent and fraction */ +#define FLOAT_PARAMS(E, F) \ + .exp_size = E, \ + .exp_bias = ((1 << E) - 1) >> 1, \ + .exp_max = (1 << E) - 1, \ + .frac_size = F, \ + .frac_shift = DECOMPOSED_BINARY_POINT - F, \ + .frac_lsb = 1ull << (DECOMPOSED_BINARY_POINT - F), \ + .frac_lsbm1 = 1ull << ((DECOMPOSED_BINARY_POINT - F) - 1), \ + .round_mask = (1ull << (DECOMPOSED_BINARY_POINT - F)) - 1, \ + .roundeven_mask = (2ull << (DECOMPOSED_BINARY_POINT - F)) - 1 + +static const FloatFmt float16_params = { + FLOAT_PARAMS(5, 10) +}; + +static const FloatFmt float32_params = { + FLOAT_PARAMS(8, 23) +}; + +static const FloatFmt float64_params = { + FLOAT_PARAMS(11, 52) +}; + /*---------------------------------------------------------------------------- | Takes a 64-bit fixed-point value `absZ' with binary point between bits 6 | and 7, and returns the properly rounded 32-bit integer corresponding to the