<addr_expr <realpart_expr <x>>> ?

[Diego Novillo]

On Sun, 2004-01-11 at 22:43, Richard Henderson wrote:
> On Sun, Jan 11, 2004 at 09:26:58PM -0500, Diego Novillo wrote:
> > How about removing COMPLEX_TYPE, COMPLEX_CST and COMPLEX_EXPR from
> > tree.def?  Have the parser turn __complex__ into 'struct complex { <t>
> > real; <t> imag }'.
> 
> No.  Various ABIs have special calling conventions for complex values
> that differ from aggregates.  Even aggregates like struct complex.
> 
Bummer.  Then we could probably move the lowering logic that Brian is
adding to SRA into lower_function_body.


Diego.