minor noises made by gcc and -std=iso9899:1999 with -Wall -pedantic -Wextra -pedantic-errors

Torbjörn Granlund tg at gmplib.org
Mon Jul 1 21:05:55 UTC 2019


Vincent Lefevre <vincent at vinc17.net> writes:

  There is nothing in the GMP manual that tells the user not to use
  -pedantic-errors (an option that should not break portable programs).

Indeed.  We do not re-document all compilers and their options'
suitability for GMP.

  Note: This option (which covers much more than empty translation units)
  can be useful for his own programs and possibly other programs, and
  might been enabled globally as a consequence.

There are many options which break programs, and if people play with
these options, they are presumably prepared to deal with the fallout.

  > I (still) believes development time is best used for solving actual
  > problem rather than hypothetical ones.  Also, if we silence pointless
  > warnings from compilers by making the code slower and/or harder to read,
  > that's not an improvement in my world.

  Not hypothetical, as I've explained just above. The development time
  of the other users is valuable too.

Absolutely!  That's why we spend so much time on actual portability and
have one of the largest test setups in the FOSS world.

-- 
Torbjörn
Please encrypt, key id 0xC8601622


More information about the gmp-discuss mailing list