Re: [PING PATCH, PR 57748] Set mode of structures with zero sized arrays to be BLK

[Richard Biener <rguenther at suse dot de>]

On Tue, 27 Aug 2013, Jakub Jelinek wrote:

> On Tue, Aug 27, 2013 at 04:03:42PM +0200, Martin Jambor wrote:
> > On Fri, Aug 23, 2013 at 05:29:23PM +0200, Jakub Jelinek wrote:
> > > On Fri, Aug 23, 2013 at 05:11:22PM +0200, Martin Jambor wrote:
> > > > Hi Jakub and/or Joseph,
> > > > 
> > > > the reporter of this bug seems to be very anxious to have it fixed in
> > > > the repository.  While Richi is away, do you think you could have a
> > > > look?  It is very small.
> > > 
> > > Isn't this ABI incompatible change (at least potential on various targets)?
> > > If so, then it shouldn't be applied to release branches, because it would
> > > create (potential?) ABI incompatibility between 4.8.[01] and 4.8.2.
> > > 
> > 
> > I don't know.  I did a few attempts to observe a change in the calling
> > convention of a function accepting a zero sized array terminated
> > structure by value (on x86_64) but I did not succeed.  On the other
> > hand, I assume there are many other ways how a mode can influence ABI.
> > So I'll try to have a look whether we can hack around this situation
> > in 4.8's expr.c instead.
> 
> All I remember that e.g. the number of regressions from PR20020 was big,
> and any kind of TYPE_MODE changes are just extremely risky.
> Perhaps x86_64 in the ABI decisions never uses mode, but we have tons of
> other targets and it would surprise me if none of those were affected.
> 
> > Nevertheless, is this patch ok for trunk?
> 
> I'll defer that to Richard now that he is back ;)

Eh ... :/

I'm extremely nervous about this change.  I also believe the change
is unrelated to the issue in the bugreport (even if it happens to
fix the ICE).

Let me have a (short) look.

Richard.