Re: PowerPC's math_ldbl.h

[Adhemerval Zanella <azanella at linux dot vnet dot ibm dot com>]

Ping.

On 02/27/2013 04:27 PM, Adhemerval Zanella wrote:
> Ping.
>
> On 02/22/2013 04:11 PM, Adhemerval Zanella wrote:
>> On 02/22/2013 12:46 PM, Joseph S. Myers wrote:
>>> On Fri, 22 Feb 2013, Adhemerval Zanella wrote:
>>>
>>>> 	* sysdeps/unix/sysv/linux/powerpc/Implies: Add powerpc/fpu prior to
>>>> 	ieee754/ldbl-128ibm so sysdeps/powerpc/fpu/math_ldbl.h will be included
>>>> 	first.
>>> That's wrong - powerpc --without-fp will be using 
>>> sysdeps/unix/sysv/linux/powerpc but can't use sysdeps/powerpc/fpu.
>>>
>> Indeed I didn't take nofpu in consideration. This patch now removes the
>> sysdeps/unix/sysv/linux/powerpc/Implies and adds the ldbl after ldbl-opt
>> on sysdeps/powerpc/Implies. This makes sysdeps/powerpc/fpu not appear with
>> nofpu build while add making it show before sysdeps/ieee754/ldbl-128ibm
>> so the PPC specific optimization is still used.
>>
>>
>> 2013-02-22  Adhemerval Zanella  <azanella@linux.vnet.ibm.com>
>>
>> 	* sysdeps/ieee754/ldbl-128ibm/math_ldbl.h (ldbl_pack): Renamed to
>> 	default_ldbl_pack and using as default implementation.
>> 	(ldbl_unpack): Renamed to default_ldbl_unpack and using as default
>> 	implementation.
>> 	* sysdeps/powerpc/fpu/math_ldbl.h (ldbl_extract_mantissa): removed
>> 	redundant definition.
>> 	(ldbl_insert_mantissa): Likewise.
>> 	(ldbl_canonicalize): Likewise.
>> 	(ldbl_nearbyint): Likewise.
>> 	(ldbl_pack): Renamed to ldbl_pack_ppc.
>> 	(ldbl_unpack): Renamed to ldbl_unpack_ppc.
>> 	* sysdeps/unix/sysv/linux/powerpc/Implies: Remove file.
>> 	* sysdeps/powerpc/Implies: Add ieee754/ldbl-opt after ieee754/ldbl-128ibm.
>>
>> --
>>
>> diff --git a/sysdeps/ieee754/ldbl-128ibm/math_ldbl.h b/sysdeps/ieee754/ldbl-128ibm/math_ldbl.h
>> index be9ac71..1cce1fc 100644
>> --- a/sysdeps/ieee754/ldbl-128ibm/math_ldbl.h
>> +++ b/sysdeps/ieee754/ldbl-128ibm/math_ldbl.h
>> @@ -125,7 +125,7 @@ ldbl_insert_mantissa (int sign, int exp, int64_t hi64, u_int64_t lo64)
>>  /* Handy utility functions to pack/unpack/cononicalize and find the nearbyint
>>     of long double implemented as double double.  */
>>  static inline long double
>> -ldbl_pack (double a, double aa)
>> +default_ldbl_pack (double a, double aa)
>>  {
>>    union ibm_extended_long_double u;
>>    u.dd[0] = a;
>> @@ -134,7 +134,7 @@ ldbl_pack (double a, double aa)
>>  }
>>
>>  static inline void
>> -ldbl_unpack (long double l, double *a, double *aa)
>> +default_ldbl_unpack (long double l, double *a, double *aa)
>>  {
>>    union ibm_extended_long_double u;
>>    u.d = l;
>> @@ -142,6 +142,12 @@ ldbl_unpack (long double l, double *a, double *aa)
>>    *aa = u.dd[1];
>>  }
>>
>> +#ifndef ldbl_pack
>> +# define ldbl_pack   default_ldbl_pack
>> +#endif
>> +#ifndef ldbl_unpack
>> +# define ldbl_unpack default_ldbl_unpack
>> +#endif
>>
>>  /* Convert a finite long double to canonical form.
>>     Does not handle +/-Inf properly.  */
>> diff --git a/sysdeps/powerpc/Implies b/sysdeps/powerpc/Implies
>> index 7ccf9a7..78dba95 100644
>> --- a/sysdeps/powerpc/Implies
>> +++ b/sysdeps/powerpc/Implies
>> @@ -1,4 +1,5 @@
>>  # On PowerPC we use the IBM extended long double format.
>>  ieee754/ldbl-128ibm
>> +ieee754/ldbl-opt
>>  ieee754/dbl-64
>>  ieee754/flt-32
>> diff --git a/sysdeps/powerpc/fpu/math_ldbl.h b/sysdeps/powerpc/fpu/math_ldbl.h
>> index 20224e6..36378c0 100644
>> --- a/sysdeps/powerpc/fpu/math_ldbl.h
>> +++ b/sysdeps/powerpc/fpu/math_ldbl.h
>> @@ -2,132 +2,12 @@
>>  #error "Never use <math_ldbl.h> directly; include <math_private.h> instead."
>>  #endif
>>
>> -#include <sysdeps/ieee754/ldbl-128/math_ldbl.h>
>> -#include <ieee754.h>
>> -  
>> -static inline void
>> -ldbl_extract_mantissa (int64_t *hi64, u_int64_t *lo64, int *exp, long double x)
>> -{
>> -  /* We have 105 bits of mantissa plus one implicit digit.  Since
>> -     106 bits are representable we use the first implicit digit for
>> -     the number before the decimal point and the second implicit bit
>> -     as bit 53 of the mantissa.  */
>> -  unsigned long long hi, lo;
>> -  int ediff;
>> -  union ibm_extended_long_double eldbl;
>> -  eldbl.d = x;
>> -  *exp = eldbl.ieee.exponent - IBM_EXTENDED_LONG_DOUBLE_BIAS;
>> -
>> -  lo = ((long long)eldbl.ieee.mantissa2 << 32) | eldbl.ieee.mantissa3;
>> -  hi = ((long long)eldbl.ieee.mantissa0 << 32) | eldbl.ieee.mantissa1;
>> -  /* If the lower double is not a denomal or zero then set the hidden
>> -     53rd bit.  */
>> -  if (eldbl.ieee.exponent2 > 0x001)
>> -    {
>> -      lo |= (1ULL << 52);
>> -      lo = lo << 7; /* pre-shift lo to match ieee854.  */
>> -      /* The lower double is normalized separately from the upper.  We
>> -	 may need to adjust the lower mantissa to reflect this.  */
>> -      ediff = eldbl.ieee.exponent - eldbl.ieee.exponent2;
>> -      if (ediff > 53)
>> -	lo = lo >> (ediff-53);
>> -    }
>> -  hi |= (1ULL << 52);
>> -  
>> -  if ((eldbl.ieee.negative != eldbl.ieee.negative2)
>> -      && ((eldbl.ieee.exponent2 != 0) && (lo != 0LL)))
>> -    {
>> -      hi--;
>> -      lo = (1ULL << 60) - lo;
>> -      if (hi < (1ULL << 52))
>> -	{
>> -	  /* we have a borrow from the hidden bit, so shift left 1.  */
>> -	  hi = (hi << 1) | (lo >> 59);
>> -	  lo = 0xfffffffffffffffLL & (lo << 1);
>> -	  *exp = *exp - 1;
>> -	}
>> -    }
>> -  *lo64 = (hi << 60) | lo;
>> -  *hi64 = hi >> 4;
>> -}
>> -
>> -static inline long double
>> -ldbl_insert_mantissa (int sign, int exp, int64_t hi64, u_int64_t lo64)
>> -{
>> -  union ibm_extended_long_double u;
>> -  unsigned long hidden2, lzcount;
>> -  unsigned long long hi, lo;
>> -
>> -  u.ieee.negative = sign;
>> -  u.ieee.negative2 = sign;
>> -  u.ieee.exponent = exp + IBM_EXTENDED_LONG_DOUBLE_BIAS;
>> -  u.ieee.exponent2 = exp-53 + IBM_EXTENDED_LONG_DOUBLE_BIAS;
>> -  /* Expect 113 bits (112 bits + hidden) right justified in two longs.
>> -     The low order 53 bits (52 + hidden) go into the lower double */ 
>> -  lo = (lo64 >> 7)& ((1ULL << 53) - 1);
>> -  hidden2 = (lo64 >> 59) &  1ULL;
>> -  /* The high order 53 bits (52 + hidden) go into the upper double */
>> -  hi = (lo64 >> 60) & ((1ULL << 11) - 1);
>> -  hi |= (hi64 << 4);
>> -
>> -  if (lo != 0LL)
>> -    {
>> -      /* hidden2 bit of low double controls rounding of the high double.
>> -	 If hidden2 is '1' then round up hi and adjust lo (2nd mantissa)
>> -	 plus change the sign of the low double to compensate.  */
>> -      if (hidden2)
>> -	{
>> -	  hi++;
>> -	  u.ieee.negative2 = !sign;
>> -	  lo = (1ULL << 53) - lo;
>> -	}
>> -      /* The hidden bit of the lo mantissa is zero so we need to
>> -	 normalize the it for the low double.  Shift it left until the
>> -	 hidden bit is '1' then adjust the 2nd exponent accordingly.  */ 
>> -
>> -      if (sizeof (lo) == sizeof (long))
>> -	lzcount = __builtin_clzl (lo);
>> -      else if ((lo >> 32) != 0)
>> -	lzcount = __builtin_clzl ((long) (lo >> 32));
>> -      else
>> -	lzcount = __builtin_clzl ((long) lo) + 32;
>> -      lzcount = lzcount - 11;
>> -      if (lzcount > 0)
>> -	{
>> -	  int expnt2 = u.ieee.exponent2 - lzcount;
>> -	  if (expnt2 >= 1)
>> -	    {
>> -	      /* Not denormal.  Normalize and set low exponent.  */
>> -	      lo = lo << lzcount;
>> -	      u.ieee.exponent2 = expnt2;
>> -	    }
>> -	  else
>> -	    {
>> -	      /* Is denormal.  */
>> -	      lo = lo << (lzcount + expnt2);
>> -	      u.ieee.exponent2 = 0;
>> -	    }
>> -	}
>> -    }
>> -  else
>> -    {
>> -      u.ieee.negative2 = 0;
>> -      u.ieee.exponent2 = 0;
>> -    }
>> -
>> -  u.ieee.mantissa3 = lo & ((1ULL << 32) - 1);
>> -  u.ieee.mantissa2 = (lo >> 32) & ((1ULL << 20) - 1);
>> -  u.ieee.mantissa1 = hi & ((1ULL << 32) - 1);
>> -  u.ieee.mantissa0 = (hi >> 32) & ((1ULL << 20) - 1);
>> -  return u.d;
>> -}
>> -  
>> -/* gcc generates disgusting code to pack and unpack long doubles.
>> -   This tells gcc that pack/unpack is really a nop.  We use fr1/fr2
>> -   because those are the regs used to pass/return a single
>> -   long double arg.  */
>> +/* GCC does not optimize the default ldbl_pack code to not spill register
>> +   in the stack. The following optimization tells gcc that pack/unpack
>> +   is really a nop.  We use fr1/fr2 because those are the regs used to
>> +   pass/return a single long double arg.  */
>>  static inline long double
>> -ldbl_pack (double a, double aa)
>> +ldbl_pack_ppc (double a, double aa)
>>  {
>>    register long double x __asm__ ("fr1");
>>    register double xh __asm__ ("fr1");
>> @@ -139,7 +19,7 @@ ldbl_pack (double a, double aa)
>>  }
>>
>>  static inline void
>> -ldbl_unpack (long double l, double *a, double *aa)
>> +ldbl_unpack_ppc (long double l, double *a, double *aa)
>>  {
>>    register long double x __asm__ ("fr1");
>>    register double xh __asm__ ("fr1");
>> @@ -150,40 +30,7 @@ ldbl_unpack (long double l, double *a, double *aa)
>>    *aa = xl;
>>  }
>>
>> +#define ldbl_pack   ldbl_pack_ppc
>> +#define ldbl_unpack ldbl_unpack_ppc
>>
>> -/* Convert a finite long double to canonical form.
>> -   Does not handle +/-Inf properly.  */
>> -static inline void
>> -ldbl_canonicalize (double *a, double *aa)
>> -{
>> -  double xh, xl;
>> -
>> -  xh = *a + *aa;
>> -  xl = (*a - xh) + *aa;
>> -  *a = xh;
>> -  *aa = xl;
>> -}
>> -
>> -/* Simple inline nearbyint (double) function .
>> -   Only works in the default rounding mode
>> -   but is useful in long double rounding functions.  */
>> -static inline double
>> -ldbl_nearbyint (double a)
>> -{
>> -  double two52 = 0x10000000000000LL;
>> -
>> -  if (__builtin_expect ((__builtin_fabs (a) < two52), 1))
>> -    {
>> -      if (__builtin_expect ((a > 0.0), 1))
>> -	{
>> -	  a += two52;
>> -	  a -= two52;
>> -	}
>> -      else if (__builtin_expect ((a < 0.0), 1))
>> -	{
>> -	  a = two52 - a;
>> -	  a = -(a - two52);
>> -	}
>> -    }
>> -  return a;
>> -}
>> +#include <sysdeps/ieee754/ldbl-128ibm/math_ldbl.h>
>> diff --git a/sysdeps/unix/sysv/linux/powerpc/Implies b/sysdeps/unix/sysv/linux/powerpc/Implies
>> deleted file mode 100644
>> index ff27cdb..0000000
>> --- a/sysdeps/unix/sysv/linux/powerpc/Implies
>> +++ /dev/null
>> @@ -1,4 +0,0 @@
>> -# Make sure these routines come before ldbl-opt.
>> -ieee754/ldbl-128ibm
>> -# These supply the ABI compatibility for when long double was double.
>> -ieee754/ldbl-opt