This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [tree-ssa] fix latent & real bugs in BIT_FIELD_REFs
On Wed, 2004-03-24 at 08:41, Paolo Bonzini wrote:
> No, it can because get_virtual_var is called whenever a non-register is
> found. BIT_FIELD_REFs themselves are not registers, but their first operand
> can be a register.
>
However, we should *never* be calling add_stmt_operand with a
BIT_FIELD_REF. BIT_FIELD_REFs are different from ARRAY_REFs and
COMPONENT_REFs in that respect. Arrays and structures are always
virtual operands. OTOH, operands to BIT_FIELD_REFs may or may not be
virtual.
Handling them together inside get_expr_operands was a mistake, after
all. Sorry I didn't notice that before. Notice how the handler for
ARRAY_REF/COMPONENT_REF calls add_stmt_operand directly, but the
original handler for BIT_FIELD_REF recurses into TREE_OPERAND(t,0).
That's why you were running into the verifier ICE.
In fact, we don't even need to handle BIT_FIELD_REFs in
get_virtual_var. It's only called from add_stmt_operand, which should
never be receiving a BIT_FIELD_REF. Another #if defined ENABLE_CHECKING
candidate.
Diego.