This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [Patch] Remove gccbug from bugreport.texi, take 3

> If -save-temps works for g77, the file won't be .i (C) or .ii (C++) and we
> should appropriately state what the suffix is.

OK, I tried it. The output is a file with .f. That should be documented in 
bugs.html then.

>  I don't know what INCLUDE
> directives do but they don't look like something that goes through the C
> preprocessor.

Apparently not. I tried that too, but the preprocessor didn't touch the 
INCLUDE line. Instead I found this little gem in f77.texi:

== The depth of nested @code{INCLUDE} references depends on
== the implementation, but typically is a positive integer.


>  The Fortran maintainer should be able to advise on what
> bugs.html ought to have in the way of Fortran-specific information to
> replace this text in the manual.

Toon, that's for you now: in the patch earlier in this thread, I try to 
throw out most of the bug reporting instructions from the manuals, and 
just refer to bugs.html. The only things fortran-wise that don't seem to 
be covered there are
a) the extension of preprocessed fortran sources
b) what to do with INCLUDEd files

Point a) seems to be easily addressed by just adding .f to the list of 
extensions of preprocessed files (is this always right, i.e. independent 
of whether .F, .fpp, .r, ... was used before preprocessing?), but for b) 
we'd need another paragraph, I guess. What's you opinion?


Wolfgang Bangerth              email:  

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]