Re: [patch] Fix PR middle-end/78468

[Jeff Law <law at redhat dot com>]

On 01/27/2017 06:43 AM, Bernd Schmidt wrote:
> On 01/27/2017 01:02 PM, Eric Botcazou wrote:
> > The attached
> > patch is a middle ground between the previously working and currently
> > broken
> > situations: if the back-end defines STACK_DYNAMIC_OFFSET, then the
> > middle-end
> > assumes that STACK_DYNAMIC_OFFSET maintains the alignment; if it doesn't,
> > which means the default value of STACK_DYNAMIC_OFFSET computed in
> > function.c
> > is used instead, then the middle-end maintains the alignment itself.
>
> I'd say let's not have a middle ground - this stuff is sufficiently
> brain-twisting that I'd rather go back to a known working state. If
> there was an error in the previous patch, let's roll it back until we
> fully understand the situation.
I tend to agree.  We've been round and round on this issue.  The code is 
a horrid mess IMHO.  Every time I think I understand it I'm proven wrong.

IIRC what motivated all this in the beginning was to eliminate 
unnecessary stack utilization.  So reverting back (assuming we can find 
all the dependent patches) should be the safe thing to do.


JEff