[llvm-dev] Floating point semantic modes

Ristow, Warren via llvm-dev llvm-dev at lists.llvm.org
Mon Jan 27 18:12:54 PST 2020


Hi,

> I'm trying to put together a set of rules for how the various floating point semantic modes should be handled in clang.

Great to see this.  Thanks!

Regarding the aspect:

> =========================
> Code-visible identifiers
> =========================
> __FAST_MATH__
>
> This symbol will only be defined if and only if all of the following are set (before pragmas are applied):
>   except_behavior { ignore }
>   fenv_access { off }
>   rounding_mode { tonearest }
>   contract { fast }
>   denormal_fp_math { PreserveSign }
>   denormal_fp32_math { PreserveSign }
>   support_math_errno { off }
>   no_honor_nans { on }
>   no_honor_infinities { on }
>   no_signed_zeros { on }
>   allow_reciprocal { on }
>   allow_approximate_fns { on }
>   allow_reassociation { on }

This makes logical sense to me.  That said, it's different than what GCC does.  GCC defines __FAST_MATH__ irrespective of the settings of:

  contract (-ffp-contract={on|off|fast})
  allow_reassociation (-f[no-]associative-math)
  allow_reciprocal (-f[no-]reciprocal-math)
  rounding_mode (-f[no-]rounding-math)

(See https://reviews.llvm.org/D72675#1829810.)

I had previously thought we should be compatible with GCC (unless we find that GCC's behavior is viewed by them as a bug).  But after reading this, the simple rule of defining __FAST_MATH__ when all of the collective flags that -ffast-math defines are set appropriately, makes more sense to me.  In any case, I thought we should overtly note the difference here, in case anyone objects.

Thanks,
-Warren Ristow


From: Kaylor, Andrew <andrew.kaylor at intel.com>
Sent: Monday, January 27, 2020 3:24 PM
To: cfe-dev at lists.llvm.org; LLVM Developers Mailing List <llvm-dev at lists.llvm.org>; Ristow, Warren <warren.ristow at sony.com>; Ulrich Weigand (Ulrich.Weigand at de.ibm.com) <Ulrich.Weigand at de.ibm.com>; Cameron McInally <cameron.mcinally at nyu.edu>; Kevin Neal <Kevin.Neal at sas.com>; Blower, Melanie I <melanie.blower at intel.com>; sepavloff at gmail.com; hfinkel at anl.gov; Sanjay Patel <spatel at rotateright.com>; Wang, Pengfei <pengfei.wang at intel.com>
Subject: Floating point semantic modes

Hi all,

I'm trying to put together a set of rules for how the various floating point semantic modes should be handled in clang. A lot of this information will be relevant to other front ends, but the details are necessarily bound to a front end implementation so I'm framing the discussion here in terms of clang. Other front ends can choose to follow clang or not. The existence of this set of semantics is an LLVM property that applies to all front ends, but the front ends will have to do something to initialize them.

I will eventually do something to convert this into an RST document and find a home for it in the clang documentation, but I'd like to start by getting input on whether everyone agrees with my judgment on how these things should work and whether I've missed anything.

Here's what I've got.

======================
FP semantic modes
======================
except_behavior { ignore, strict, may_trap }
fenv_access { on, off }
rounding_mode { dynamic, tonearest, downward, upward, towardzero }
contract { on, off, fast }
denormal_fp_math { IEEE, PreserveSign, PositiveZero }
denormal_fp32_math { IEEE, PreserveSign, PositiveZero }
support_math_errno { on, off }
no_honor_nans { on, off }
no_honor_infinities { on, off }
no_signed_zeros { on, off }
allow_reciprocal { on, off }
allow_approximate_fns { on, off }
allow_reassociation { on, off }

---------------------
Dependencies
---------------------
rounding_mode must be "tonearest" if fenv_access is "off" -- It is the user's responsibility to make sure the dynamic rounding mode is not left in another state.

if except_behavior is not "ignore", allow_reciprocal, allow_approximate_fns and allow_reassociation must be "off".

======================
FP models
======================
-----------------------
precise (default)
-----------------------
except_behavior { ignore }
fenv_access { off }
rounding_mode { tonearest }
contract { on }
denormal_fp_math { IEEE }
denormal_fp32_math { IEEE }
support_math_errno { on }
no_honor_nans { off }
no_honor_infinities { off }
no_signed_zeros { off }
allow_reciprocal { off }
allow_approximate_fns { off }
allow_reassociation { off }

------------------
strict
------------------
except_behavior { strict }
fenv_access { on }
rounding_mode { dynamic }
contract { off }
denormal_fp_math { IEEE }
denormal_fp32_math { IEEE }
support_math_errno { on }
no_honor_nans { off }
no_honor_infinities { off }
no_signed_zeros { off }
allow_reciprocal { off }
allow_approximate_fns { off }
allow_reassociation { off }

------------------
fast
------------------
except_behavior { ignore }
fenv_access { off }
rounding_mode { tonearest }
contract { fast }
denormal_fp_math { PreserveSign }
denormal_fp32_math { PreserveSign }
support_math_errno { off }
no_honor_nans { on }
no_honor_infinities { on }
no_signed_zeros { on }
allow_reciprocal { on }
allow_approximate_fns { on }
allow_reassociation { on }

======================
Command-line options
======================
-ffp-model={precise|strict|fast}
  Sets all semantic modes as described above.

-ffast-math
  Equivalent to -ffp-model=fast. (I'm not sure that's currently true.)

-f[no-]math-errno
-ffp-contract={on|off|fast}
-f[no-]honor-infinities
-f[no-]honor-nans
-f[no-]associative-math
-f[no-]reciprocal-math
-f[no-]signed-zeros
-f[no-]trapping-math
-f[no-]rounding-math
-fdenormal-fp-math={ieee, preservesign, positivezero}
-fdenormal-fp-math-fp32={ieee, preservesign, positivezero}
-ffp-exception-behavior={ignore,maytrap,strict}
  Each of these has a 1-to-1 correspondance to an FP semantic mode.
  (I think several of these should set "except_behavior" to "ignore".)


-f[no-]finite-math-only
  Controls no_honor_nans and no_honor_infinities.

-f[no-]unsafe-math-optimizations
  Turns no_signed_zeros, allow_reciprocal, allow_approximate_fns, and allow_reassociation on or off.
  Also, sets except_behavior to "on" for -funsafe-math-optimizations.
  (Currently, -fno-]unsafe-math-optimizations clears except_behavior, but I regard this as a bug.)

All command line options will override any previous values of all settings they control with options taking effect in a left-to-right manner.

======================
pragmas
======================
STDC FENV_ACCESS {ON|OFF}
  Patch in progress. I think ON should force the following:

    except_behavior { strict }
    fenv_access { on }
    rounding_mode { dynamic }
    denormal_fp_math { IEEE }
    denormal_fp32_math { IEEE }
    no_signed_zeros { off }
    allow_reciprocal { off }
    allow_approximate_fns { off }
    allow_reassociation { off }

  And OFF should set fenv_access to off, except_behavior to ignore, and rounding_mode to tonearest. Other modes should be reset to their command line defined settings.

  I don't think this pragma should have any effect on contract, support_math_errno, no_honor_nans, or no_honor_infinities.

STDC FP_CONTRACT {ON|OFF|DEFAULT}
  This pragma controls the contract FP semantic mode. No other FP semantic modes are effected.

float_control ({precise|except}, {on|off}[, push])
float_control (pop)
  Patch in progress. These are tricky.
  I think they should have the following effects:

float_control (precise, on[, push])
  contract { on }
  denormal_fp_math { IEEE }
  denormal_fp32_math { IEEE }
  no_signed_zeros { off }
  allow_reciprocal { off }
  allow_approximate_fns { off }
  allow_reassociation { off }

float_control (precise, off[, push])
  contract { fast }
  denormal_fp_math { preservesign }
  denormal_fp32_math { preservesign }
  no_signed_zeros { on }
  allow_reciprocal { on }
  allow_approximate_fns { on }
  allow_reassociation { on }

Note, this is less than what the -ffp-model=precise control does. Should this override support_math_errno, no_honor_nans, or no_honor_infinities?

float_control (except, on[, push])
  except_behavior { strict }

float_control (except, off[, push])
  except_behavior { ignore }

The MSVC documentation says you can only use the float_control pragma to turn exception semantics on when precise semantics are enabled. For us, this would mean:
  denormal_fp_math { IEEE }
  denormal_fp32_math { IEEE }
  no_signed_zeros { off }
  allow_reciprocal { off }
  allow_approximate_fns { off }
  allow_reassociation { off }

The MSVC documentation also says you can't use the float_control pragma to turn excpetion semantics off when precise semantics are enabled, and you can't use the float_control pragma to turn precise off when fenv_access is on.

I believe we should follow the MSVC restrictions.

=========================
Code-visible identifiers
=========================
__FAST_MATH__

This symbol will only be defined if and only if all of the following are set (before pragmas are applied):
  except_behavior { ignore }
  fenv_access { off }
  rounding_mode { tonearest }
  contract { fast }
  denormal_fp_math { PreserveSign }
  denormal_fp32_math { PreserveSign }
  support_math_errno { off }
  no_honor_nans { on }
  no_honor_infinities { on }
  no_signed_zeros { on }
  allow_reciprocal { on }
  allow_approximate_fns { on }
  allow_reassociation { on }

__FINITE_MATH_ONLY__

This symbol will only be defined if and only if all of the following are set (before pragmas are applied):
  no_honor_nans { on }
  no_honor_infinities { on }

FLT_ROUNDS

Should be set to -1 (indeterminable) if rounding_mode() is dynamic or 1 (tonearest) if rounding_mode is tonearest. There are values for other rounding modes, but clang offers no way to set those rounding modes.

FLT_EVAL_METHOD

Should be set to -1 if any of allow_reciprocal, allow_approximate_fns, or allow_reassociation is set. Should any other flags also make this -1? Otherwise, the setting is target-defined.

math_errhandling

The MATH_ERRNO bit will be set or cleared based on the setting of support_math_errno. Should MATH_ERREXCEPT be set or cleared based on except_behavior?


Thanks in advance for any opinions and suggestions.

-Andy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20200128/81f07ca5/attachment.html>


More information about the llvm-dev mailing list