This is the mail archive of the gcc-patches@gcc.gnu.org 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: [PR58315] reset inlined debug vars at return-to point


On Feb 25, 2015, Alexandre Oliva <aoliva@redhat.com> wrote:

> if a function is called within a loop and we inline it, bindings from
> the call in one iteration of the loop will carry over onto the
> subsequent iteration until a new binding occurs.

Wait, I have to take that back and revisit the code I looked at.
var-tracking handles confluences differently between VTA-tracked and
old-style var-tracking vars, and I was probably confused by that: it
occurred to me this morning that confluence points of VTA-tracked
variables perform a set intersection, whereas old-style VTA performs set
union across locations at all incoming edges.  So, what I have concluded
should only apply to old-style var-tracking vars, or to VTA-tracked vars
in unrolled loops whose intermediate conditions were optimized out (so
that there isn't any confluence with incoming edges without bindings for
the var).

I'm afraid I'll have to look again and figure out what I got wrong.

Apologies for the noise.

-- 
Alexandre Oliva, freedom fighter    http://FSFLA.org/~lxoliva/
You must be the change you wish to see in the world. -- Gandhi
Be Free! -- http://FSFLA.org/   FSF Latin America board member
Free Software Evangelist|Red Hat Brasil GNU Toolchain Engineer


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