https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93727

--- Comment #11 from Jerry DeLisle <jvdelisle at gcc dot gnu.org> ---
Created attachment 62649
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62649&action=edit
Preliminary patch for testing

This attached patch implements the EX format specifier for WRITE. Remaining to
be done is the quadmath portion. I am considering some factoring of portions of
the write_ex function in write.c.

The following compiles and runs:

program main
  implicit none
  character s1  
  real(4) :: r4
  real(8) :: r8
  real(10) :: r10
  !real(16) :: r16  
  r4 = -huge(1.0_4/3.0_4)
  r8 = -huge( 1.0_8/3.0_8)
  r10= -huge(1.0_10/3.0_10)
  !r16 = 1.0_16/3.0_16

  write(*,"(A)") "123456789012345678901234567890"
  write(*,"(EX0.0)") r4
  write(*,"(EX0.0)") r8
  write(*,"(EX0.0)") r10
  write(*,"(A)") "123456789012345678901234567890"
  write(*,"(EX0.0)") 1.0_4/r4
  write(*,"(EX0.0)") 1.0_8/r8
  write(*,"(EX0.0)") 1.0_10/r10
  write(*,"(A)") "123456789012345678901234567890"
  !write(*,"(,EX36.4)") r16
end program main

Reply via email to