This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Ada subtypes and base types
On 03/31/06 04:08, Jeffrey A Law wrote:
> Then we just need to make VRP not be so quick to ignore statements
> with virtual operands (ie, the same problem we need to solve if
> VRP is going to be able to use non-null attributes on function
> return values).
>
Certainly. The approach I had sketched out earlier should be able to
accomodate all this:
- The front-end encodes in a langhook the notion of non-null attribute
for symbols or function return values or types.
- Propagators propagate that attribute via the propagate_value helper.
This attribute becomes an SSA name attribute (we need flow sensitivity),
so we will need to be able to encode various attributes on SSA names
(perhaps using the aux field).
- VRP (and others) can pick up those propagated attributes when doing
their thing.