Jan Stolarek

I measured compile times with GHC 7.10 and size of generated Core for `zipWith` and `uznip` functions. It's hard to say what the growth function for compile times is. It...

I did the measurements for GHC 7.8.4 and the results are not at all what I expected. Compile times (in seconds): | | 7.10.1 | 7.8.4 | | --- |...

Some more numbers: `zipWith` - bytes allocated in the heap | | 7.10.1 | 7.8.4 | | --- | --- | --- | | zipWith | 422,458,240 | 482,987,584 |...

As another tests I checked how much call arity analysis affects compilation time. I compiled Zips module using GHC 7.10.1 with `-O` and `-O -fno-call-arity`. Compilation time went down from...

Now that match flattening is reverted the problem is gone. Closing in favour of #96.

This test looks like this: ``` haskell $(singletons [d| data Nat where Zero :: Nat Succ :: Nat -> Nat deriving (Eq, Show, Read) plus :: Nat -> Nat ->...

``` Singletons/Nat.hs:0:0: Looking up Succ_0123456789 in the list of available declarations failed. This lookup fails if the declaration referenced was made in the same Template Haskell splice as the use...

In that case I'm reverting my earlier patch.

This problem has gone away after reverting match flattening.