#2132: Optimise nested comparisons
---------------------------------+------------------------------------------
Reporter: simonpj | Owner: simonpj
Type: bug | Status: patch
Priority: low | Milestone: 7.4.1
Component: Compiler | Version: 6.8.2
Keywords: | Testcase:
Blockedby: | Difficulty: Unknown
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: Runtime performance bug
---------------------------------+------------------------------------------
Changes (by basvandijk):
* cc: v.dijk.bas@… (added)
Comment:
Would it be hard to extend the language of RULES so that a user can
perform this optimization instead of the compiler? I'm thinking about
something like this:
{{{
{-# RULES
forall x y t f.
(case (x ># y) of
True -> t
False -> f
)
=
(case (x ># y) of
True -> t {{(x ==# y)=False}}
False -> f
)
#-}
}}}
The `v {{subrules}}` syntax assigns ''sub-RULES'' to the variable `v`.
`subrules` is a list of RULES of the form `p1 = e1, p2 = e2 ... pn = en`
which are applied to the expression that `v` refers to.
I think it would be nice if sub-RULES have the same (or a very similar)
syntax that normal RULES have. They should even be able to quantify new
variables.
The way I expect this to be used most often is that the sub-RULES describe
the actual substitutions while the ''top-level'' RULE only describes
context. However it should be possible that both the top-level and the
sub-rules describe transformations.
This probably deserves its own ticket so I stop now.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/2132#comment:17>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
