[PATCH] Fix readdir_r with long file names

[Florian Weimer]

On 03/02/2016 06:44 PM, Paul Eggert wrote:
> +/* Declare a structure member named ID, with type "array of TYPE", and
> +   with actual size unspecified but with nominal size SIZE for the
> +   benefit of traditional applications.  If the application is
> +   compiled with _FLEXARRAY_SOURCE and if the compiler is C99 or
> +   better, model this as a flexible array.  Otherwise, model it as a
> +   fixed-size array of size SIZE (followed by a flexible array
> +   afterwards, if C99, for the benefit of source-code analyzers that
> +   can use this as a cue that the array is really varying-size).  The
> +   fixed-size array is for compatibility with programs that unwisely
> +   take the size of structures that contain flexible arrays, and which
> +   rely on traditional glibc which uses fixed-size arrays.  */
> +#ifndef __flexarray
> +# ifdef __USE_FLEXARRAY
> +#  define __flexarray(id, type, size) type id[]
> +# else
> +#  define __flexarray(id, type, size) type id[size]; type __flexarray __flexarr
> +# endif
> +#endif

Sorry, I'm not convinced this is good compiler hint.  I suspect it moves
things in the opposite direction.  Some compilers likely have hacks to
support arrays of length 1 as flexible arrays if the array is at the end
of a struct, and after your change, the member is no longer at the end,
potentially disabling the flexible array kludge.

I think we should interact better with compilers in this area, but this
needs fairly explicit mark-up, showing what we need, and not something
that requires the compiler to guess.

Thanks,
Florian