[llvm-dev] RFC: New IR attribute incoming-stack-align

[Joerg Sonnenberger via llvm-dev]

On Tue, Mar 15, 2016 at 10:48:30AM +0000, Huw Davies via llvm-dev wrote:
> An alternative would be to have the 'alignstack' IR attribute itself
> reduce the incoming stack alignment assumption.  i.e. if it's
> specified then it's fair to assume that the incoming stack does not
> have the correct alignment, so we should assume it has a smaller
> alignment - which we would probably just pick as 4.  Would that be any
> better?

Wouldn't a better way be to say that if alignstack is specified,
arguments have their minimum required alignment and that's it?

Joerg