r190196 - in /branches/cxx-conversion/gcc: Chan...

Tue Aug 7 00:38:00 GMT 2012

Author: crowl
Date: Tue Aug  7 00:38:52 2012
New Revision: 190196

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190196
Log:
Convert double_int from a struct with function into a class with
operators and methods.

This patch adds the methods and operators.  In general functions of
the form "double_int_whatever" become member functions "whatever" or,
when possible, operators.

Every attempt has been made to preserve the existing algorithms, even
at the expense of some optimization opportunities.  Some examples:

  The ext operation takes a value and returns a value.  However, that
  return value is usually assigned to the original variable.  An
  operation that modified a variable would be more efficient.

  In some cases, an outer sign-specific function calls an inner
  function with the sign as a parameter, which then decides which
  implementation to do.  Decisions should not be artificially
  introduced, and the implementation of each case should be exposed as
  a separate routine.

  The existing operations are implemented in terms of the new
  operations, which necessarily adds a layer between the new code and
  the existing users.  Once all files have migrated, this layer will
  be removed.

  There are several existing operations implemented in terms of even
  older legacy operations.  This extra layer has not been removed.

On occasion though, parameterized functions are often called
with a constant argments.  To support static statement of intent,
and potentially faster code in the future, there are several new
unparameterized member functions.  Some examples:

  Four routines now encode both 'arithmetic or logical' and 'right or
  left' shift as part of the funciton name.

  Four routines now encode both 'signed or unsigned' and 'less than or
  greater than' as part of the function name.

This patch also converts the file fixed-value.c as an example of the
kind of source level changes.  As part of the process, some of the
"grandfather" calls have been migrated.

To preserve sensible diffs, some transitional local reference
variables have been introduced into the code.  These reference
variables provide the same name for access to *this that the functions
formerly used as parameter names.  As a result, many expressions are
the same and show no diff.  Later patches can remove the transitional
reference variable.

So, this patch is an intermediate patch.  The time for the compiler to
build itself is 0.7% longer.  Most of that should be recovered when
uses of the old interface are removed and the extra layer of functions
are no longer present.  Further performance should be available from
factoring an implementation that will no longer be exposed to clients.

Tested on x86.

Index: gcc/ChangeLog.cxx-conversion

2012-08-06   Lawrence Crowl  <crowl@google.com>

	* hash-table.h
	(typedef double_int): Change to struct (POD).
	(double_int::make): New overloads for int to double-int conversion.
	(double_int::mask): New.
	(double_int::max_value): New.
	(double_int::min_value): New.
	(double_int::operator ++): New.
	(double_int::operator --): New.
	(double_int::operator *=): New.
	(double_int::operator +=): New.
	(double_int::operator -=): New.
	(double_int::to_signed): New.
	(double_int::to_unsigned): New.
	(double_int::fits_unsigned): New.
	(double_int::fits_signed): New.
	(double_int::fits): New.
	(double_int::trailing_zeros): New.
	(double_int::popcount): New.
	(double_int::multiple_of): New.
	(double_int::set_bit): New.
	(double_int::mul_with_sign): New.
	(double_int::operator * (binary)): New.
	(double_int::operator + (binary)): New.
	(double_int::operator - (binary)): New.
	(double_int::operator - (unary)): New.
	(double_int::operator ~ (unary)): New.
	(double_int::operator & (binary)): New.
	(double_int::operator | (binary)): New.
	(double_int::operator ^ (binary)): New.
	(double_int::and_not): New.
	(double_int::lshift): New.
	(double_int::rshift): New.
	(double_int::alshift): New.
	(double_int::arshift): New.
	(double_int::llshift): New.
	(double_int::lrshift): New.
	(double_int::lrotate): New.
	(double_int::rrotate): New.
	(double_int::div): New.
	(double_int::sdiv): New.
	(double_int::udiv): New.
	(double_int::mod): New.
	(double_int::smod): New.
	(double_int::umod): New.
	(double_int::divmod): New.
	(double_int::sdivmod): New.
	(double_int::udivmod): New.
	(double_int::ext): New.
	(double_int::zext): New.
	(double_int::sext): New.
	(double_int::is_zero): New.
	(double_int::is_one): New.
	(double_int::is_minus_one): New.
	(double_int::is_negative): New.
	(double_int::cmp): New.
	(double_int::ucmp): New.
	(double_int::scmp): New.
	(double_int::ult): New.
	(double_int::ugt): New.
	(double_int::slt): New.
	(double_int::sgt): New.
	(double_int::max): New.
	(double_int::smax): New.
	(double_int::umax): New.
	(double_int::min): New.
	(double_int::smin): New.
	(double_int::umin): New.
	(double_int::operator ==): New.
	(double_int::operator !=): New.
	(shwi_to_double_int): Change implementation to use member function.
	(double_int_minus_one): Likewise.
	(double_int_zero): Likewise.
	(double_int_one): Likewise.
	(double_int_two): Likewise.
	(double_int_ten): Likewise.
	(uhwi_to_double_int): Likewise.
	(double_int_to_shwi): Likewise.
	(double_int_to_uhwi): Likewise.
	(double_int_fits_in_uhwi_p): Likewise.
	(double_int_fits_in_shwi_p): Likewise.
	(double_int_fits_in_hwi_p): Likewise.
	(double_int_mul): Likewise.
	(double_int_mul_with_sign): Likewise.
	(double_int_add): Likewise.
	(double_int_sub): Likewise.
	(double_int_neg): Likewise.
	(double_int_div): Likewise.
	(double_int_sdiv): Likewise.
	(double_int_udiv): Likewise.
	(double_int_mod): Likewise.
	(double_int_smod): Likewise.
	(double_int_umod): Likewise.
	(double_int_divmod): Likewise.
	(double_int_sdivmod): Likewise.
	(double_int_udivmod): Likewise.
	(double_int_multiple_of): Likewise.
	(double_int_setbit): Likewise.
	(double_int_ctz): Likewise.
	(double_int_not): Likewise.
	(double_int_ior): Likewise.
	(double_int_and): Likewise.
	(double_int_and_not): Likewise.
	(double_int_xor): Likewise.
	(double_int_lshift): Likewise.
	(double_int_rshift): Likewise.
	(double_int_lrotate): Likewise.
	(double_int_rrotate): Likewise.
	(double_int_cmp): Likewise.
	(double_int_scmp): Likewise.
	(double_int_ucmp): Likewise.
	(double_int_max): Likewise.
	(double_int_smax): Likewise.
	(double_int_umax): Likewise.
	(double_int_min): Likewise.
	(double_int_smin): Likewise.
	(double_int_umin): Likewise.
	(double_int_ext): Likewise.
	(double_int_sext): Likewise.
	(double_int_zext): Likewise.
	(double_int_mask): Likewise.
	(double_int_max_value): Likewise.
	(double_int_min_value): Likewise.
	(double_int_zero_p): Likewise.
	(double_int_one_p): Likewise.
	(double_int_minus_one_p): Likewise.
	(double_int_equal_p): Likewise.
	(double_int_popcount): Likewise.
	* hash-table.c
	(double_int_mask): Reuse implementation for double_int::mask.
	(double_int_max_value): Likewise.
	(double_int_min_value): Likewise.
	(double_int_ext): Likewise.
	(double_int_zext): Likewise.
	(double_int_sext): Likewise.
	(double_int_mul_with_sign): Likewise.
	(double_int_divmod): Likewise.
	(double_int_sdivmod): Likewise.
	(double_int_udivmod): Likewise.
	(double_int_div): Likewise.
	(double_int_sdiv): Likewise.
	(double_int_udiv): Likewise.
	(double_int_mod): Likewise.
	(double_int_smod): Likewise.
	(double_int_umod): Likewise.
	(double_int_multiple_of): Likewise.
	(double_int_lshift): Likewise.
	(double_int_rshift): Likewise.
	(double_int_lrotate): Likewise.
	(double_int_rrotate): Likewise.
	(double_int_cmp): Likewise.
	(double_int_ucmp): Likewise.
	(double_int_scmp): Likewise.
	(double_int_max): Likewise.
	(double_int_smax): Likewise.
	(double_int_umax): Likewise.
	(double_int_min): Likewise.
	(double_int_smin): Likewise.
	(double_int_umin): Likewise.
	(double_int_min): Likewise.
	(double_int_min): Likewise.
	(double_int_min): Likewise.
	(double_int_min): Likewise.
	(double_int_min): Likewise.
	(double_int_min): Likewise.
	(double_int::alshift): New.
	(double_int::arshift): New.
	(double_int::llshift): New.
	(double_int::lrshift): New.
	(double_int::ult): New.
	(double_int::ugt): New.
	(double_int::slt): New.
	(double_int::sgt): New.
	(double_int_setbit): Reuse implementation for double_int::set_bit,
	which avoids a name conflict with a macro.
	(double_int_double_int_ctz): Reuse implementation for
	double_int::trailing_zeros.
	(double_int_fits_in_shwi_p): Reuse implementation for
	double_int::fits_signed.
	(double_int_fits_in_hwi_p): Reuse implementation for double_int::fits.
	(double_int_mul): Reuse implementation for binary
	double_int::operator *.
	(double_int_add): Likewise.
	(double_int_sub): Likewise.
	(double_int_neg): Reuse implementation for unary
	double_int::operator -.
	(double_int_max_value): Likewise.
	* fixed-value.c: Change to use member functions introduced above.

Modified:
    branches/cxx-conversion/gcc/ChangeLog.cxx-conversion
    branches/cxx-conversion/gcc/double-int.c
    branches/cxx-conversion/gcc/double-int.h
    branches/cxx-conversion/gcc/fixed-value.c