bug#37140: 27.0.50; calc gives "Wrong type argument: number-or-marker-p, inf" to many operations

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > In other words, the route to the error was somewhat roundabout,
  > but it looks like there is no actual bug in Emacs nor in gcc. But
  > at least I didn't know that, and was enlightened by the result.

Whether it is a bug in Emacs is a matter of judgment / definition.  We
could say, "Don't compile Emacs with -ffinite-math-only", in which
case this is not a bug in Emacs.  Or we could consider
-ffinite-math-only legit, in which case this is a bug in Emacs.

We might want to do the latter IF (1) this bug would be easy to fix
and (2) there wouldn't be other problems as a result.  I suspect
this one problem would be easy to fix.

If Emacs is supposed to handle infinite floats, then compiling with
-ffinite-math-only is inescapably bad because it will make the
infinite floats fail to work.

In that case, maybe we should put in something to report an error
for certain if Emacs is compiled with -ffinite-math-only.
One way would be to run some code at startup which ought to
generate an infinity. tests that it did so, and otherwise
gives a clear fatal error message.




-- 
Dr Richard Stallman
President, Free Software Foundation (https://gnu.org, https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)