This is the mail archive of the gcc@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: fold() can't fold simple expressions?


On 09/14/2016 03:29 PM, Richard Biener wrote:
On September 14, 2016 6:39:14 PM GMT+02:00, Jeff Law <law@redhat.com> wrote:
On 09/14/2016 08:08 AM, Andrew MacLeod wrote:
range generator understands, we just thought it would be handy to
leverage the folder during the proof of concept stage.
It's also worth noting that a backwards substitution based threading,
redundancy elimination, etc pass would need similar capabilities.

Essentially you start with some gimple expression, say a + b.  You then

start walking backwards substituting the RHS of defining statements
into
the expression and simplifying as you go.
It's what match-and-simplify does as well.

I question the need to build GENERIC here though.  M-a-s happily gets you a simplified expression as sequence of GIMPLE statements.  (But does not yet provide a way to build a simplified GENERIC expression from GIMPLE IL)




I'm considering alternatives and we may be able to avoid it by marking and delaying some evaluations... it was just an easy way to an ends for prototyping.

Andrew


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