#4941: SpecConstr generates functions that do not use their arguments
---------------------------------+------------------------------------------
Reporter: simonpj | Owner:
Type: task | Status: new
Priority: normal | Milestone: _|_
Component: Compiler | Version: 7.0.1
Keywords: | Testcase:
Blockedby: | Difficulty:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: Runtime performance bug
---------------------------------+------------------------------------------
Comment(by batterseapower):
Indeed. I found that you need to have a fixed point between SAT and
SpecConstr to deal with the output of stream fusion. Similarly there is an
argument for having a fixed point between SAT and strictness analysis
because it is possible that strictness analysis (in particular CPR-
optimisation) would give rise to further opportunities for SpecConstr.
Consider:
{{{
f b x = case b of A -> I# y; B -> x; C -> f A (x + 1)
}}}
After SpecConstr, the (f A z) specialisation has the CPR property, as
worker-wrapper will reveal. So if there was some other call:
{{{
g x = g (f A x)
}}}
After inlining the wrapper we now have a SpecConstr opportunity for (g (I#
z)). I wonder how often this would actually occur in practice..
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4941#comment:3>
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
