#4428: Local functions lose their unfoldings
---------------------------------+------------------------------------------
Reporter: rl | Owner:
Type: bug | Status: new
Priority: normal | Component: Compiler
Version: 7.1 | Keywords:
Testcase: | Blockedby:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: Runtime performance bug
---------------------------------+------------------------------------------
I'm attaching 2 modules, both should be compiled with -O2. Let's look at
the inteface of `Foo`:
{{{
foo :: (GHC.Types.Int, GHC.Types.Int)
-> [GHC.Types.Int]
-> [GHC.Types.Int]
{- Arity: 2, Strictness: U(U(L)U(L))L, Inline: INLINE (sat-args=2),
Unfolding: InlineRule (2, False, False)
(\ ds :: (GHC.Types.Int, GHC.Types.Int) xs ::
[GHC.Types.Int] ->
case @ [GHC.Types.Int] ds of wild { (i, n) ->
let {
step :: forall t.
([t], GHC.Types.Int, GHC.Types.Int)
-> Data.Maybe.Maybe (([t], GHC.Types.Int,
GHC.Types.Int), t)
{- Inline: INLINE[0] (sat-args=1) -}
= ...
}}}
Note how `step`, which is local to `foo`, has an `INLINE` pragma but no
unfolding. Now, let's look at `Bar`. For some reason, GHC only inlines
`foo` into `bar` in phase 1. This is probably unrelated but still
surprising. The main problem, though, is that `step`, not having an
unfolding, doesn't get inlined at all despite the `INLINE` pragma.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4428>
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
