Re: [PATCH fortran/diagnostics] Move gfc_error (buffered) to common diagnostics

[Dodji Seketeli <dodji at redhat dot com>]

Hello Manuel,

Manuel LÃpez-IbÃÃez <lopezibanez@gmail.com> writes:

> New version of the patch. Tobias noticed several problems with the
> previous version:
>
> * Due to the use of placement-new for the buffered output_buffers
> pp_warning_buffer and pp_error_buffer, the pretty-printer destructor
> may end up trying to free something that it can't. Fixed here by not
> using placement new.

So:

[...]


>  /* Report the number of warnings and errors that occurred to the caller.  */
>  
> @@ -1525,11 +1625,14 @@ gfc_diagnostics_init (void)
>  {
>    diagnostic_starter (global_dc) = gfc_diagnostic_starter;
>    diagnostic_finalizer (global_dc) = gfc_diagnostic_finalizer;
>    diagnostic_format_decoder (global_dc) = gfc_format_decoder;
>    global_dc->caret_char = '^';
> -  new (&pp_warning_buffer) output_buffer ();
> +  pp_warning_buffer = new output_buffer ();

When I look at the code of the destructor the pretty_printer type
(pretty_printer::~pretty_printer) in gcc/pretty-print.c, I see that the
memory for the output buffer is de-allocated using XDELETE.  So I think
the memory for the output buffer should be allocated using XNEW and the
output_buffer type should instantiated using a placement new operator
that uses that XNEWed allocated memory.

Ultimately, I should sit down and make sure the memory management of the
pretty_printer type does away with this surprising placement new
business for good.

> +  pp_warning_buffer->flush_p = false;
> +  pp_error_buffer = new output_buffer ();
> +  pp_error_buffer->flush_p = false;
>  }

Cheers,

-- 
		Dodji