tilelang.contrib.cutedsl.ieee_math
==================================

.. py:module:: tilelang.contrib.cutedsl.ieee_math

.. autoapi-nested-parse::

   IEEE-754 compliant floating-point operations with explicit rounding modes.

   These correspond to CUDA __fadd_rn, __fsub_rz, etc. Implemented via inline PTX
   to ensure exact rounding mode compliance.

   Rounding modes: rn (nearest), rz (toward zero), rm (toward -inf), rp (toward +inf)



Functions
---------

.. autoapisummary::

   tilelang.contrib.cutedsl.ieee_math.ieee_fadd
   tilelang.contrib.cutedsl.ieee_math.ieee_fsub
   tilelang.contrib.cutedsl.ieee_math.ieee_fmul
   tilelang.contrib.cutedsl.ieee_math.ieee_fmaf
   tilelang.contrib.cutedsl.ieee_math.ieee_frcp
   tilelang.contrib.cutedsl.ieee_math.ieee_fsqrt
   tilelang.contrib.cutedsl.ieee_math.ieee_fdiv


Module Contents
---------------

.. py:function:: ieee_fadd(a, b, rounding='rn')

   IEEE-754 add with explicit rounding mode.


.. py:function:: ieee_fsub(a, b, rounding='rn')

   IEEE-754 subtract with explicit rounding mode.


.. py:function:: ieee_fmul(a, b, rounding='rn')

   IEEE-754 multiply with explicit rounding mode.


.. py:function:: ieee_fmaf(a, b, c, rounding='rn')

   IEEE-754 fused multiply-add with explicit rounding mode.


.. py:function:: ieee_frcp(a, rounding='rn')

   IEEE-754 reciprocal with explicit rounding mode.


.. py:function:: ieee_fsqrt(a, rounding='rn')

   IEEE-754 square root with explicit rounding mode.


.. py:function:: ieee_fdiv(a, b, rounding='rn')

   IEEE-754 divide with explicit rounding mode.