This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [Patch, fortran] PR20541 - allocatable components (TR15541)

On Sun, Oct 08, 2006 at 01:14:47PM +0200, Paul Thomas wrote:
> >>Out of curiousity, why didn't you use alloca instead of doing all 
> >>this  bookkeeping?  Maybe because you would have needed bookkeeping 
> >>anyway  for the case where the allocatable component would have been 
> >>>  -fmax-stack-var-size?  Or is there a simpler reason?
> >
> >Ignorance. 
> I was going back through the correspondence, getting ready for tomorrow 
> morning's commit, when I saw this.
> I was speaking for myself here! Erik might well have known all about it 
> and had cogent reasons for not using it.

I knew of alloca, but my personal experiences with it (in other
projects than GCC/gfortran) have been that it isn't portable, and I have
therefore developed an almost instinctual dislike for it.  I wasn't
aware of __builtin_alloca, which of course ought be OK (from portability
point of view; there's of course still the limited-stack-size issue to
worry about).


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]