Patch: scheduling vs variable arrays

[Richard Henderson]

On Thu, Aug 08, 2002 at 04:56:38PM -0700, Geoff Keating wrote:
> - I think it would be better to use (clobber (mem:BLK (reg:P sp)))
>   because that makes it clear that only stack memory is involved.

Except that only (mem:blk (scratch)) is special-cased _before_
the rtx_unchanging_p check.  And if for some reason you declare
a variable-sized array as constant (with initializer, presumably),
then it will be rtx_unchanging_p.

> - Shouldn't the clobber be attached to an insn rather than floating
>   around loose?  For instance, the stack pop insn?  Otherwise it's
>   possible that while the stores won't be moved below the stack pop,
>   the stack pop might be moved above the stores, which will have
>   pretty much the same effect.
>
> For rs6000, we already have stack_tie_insn which is used for this
> purpose, see rs6000_emit_stack_tie.

For this purpose, we'd have to add a named insn to every backend.


r~