diff options
author | Simon Peyton Jones <simonpj@microsoft.com> | 2015-04-29 13:43:09 +0100 |
---|---|---|
committer | Simon Peyton Jones <simonpj@microsoft.com> | 2015-04-29 15:05:44 +0100 |
commit | a1275a762ec04c1159ae37199b1c8f998a5c5499 (patch) | |
tree | 6d01de796910a7857c5f91661a40592f67d62090 /testsuite/tests/module | |
parent | d9bb0ee78b2104bd5c71508080d068804086ae84 (diff) | |
download | haskell-a1275a762ec04c1159ae37199b1c8f998a5c5499.tar.gz |
Improve improvement in the constraint solver
This regrettably-big patch substantially improves the way in which
"improvement" happens in the constraint solver. It was triggered by
trying to crack Trac #10009, but it turned out to solve #10340 as
well.
The big picture, with several of the trickiest examples, is described
in Note [The improvement story] in TcInteract.
The major change is this:
* After solving we explicitly try "improvement", by
- making the unsolved Wanteds into Deriveds
- allowing Deriveds to rewrite Deriveds
This more aggressive rewriting "unlocks" some extra
guess-free unifications.
* The main loop is in TcInteract.solveSimpleWanteds, but I also ended
up refactoring TcSimplify.simpl_loop, and its surrounding code.
Notably, any insolubles from the Givens are pulled out
and treated separately, rather than staying in the inert set
during the solveSimpleWanteds loop.
There are a lot of follow-on changes
* Do not emit generate Derived improvements from Wanteds.
This saves work in the common case where they aren't needed.
* For improvement we should really do type-class reduction on Derived
constraints in doTopReactDict. That entailed changing the GenInst
constructor a bit; a local and minor change
* Some annoying faffing about with dropping derived constraints;
see dropDerivedWC, dropDerivedSimples, dropDerivedInsols,
and their Notes.
* Some substantial refactoring in TcErrors.reportWanteds.
This work wasn't strictly forced, but I got sucked into it.
All the changes are in TcErrors.
* Use TcS.unifyTyVar consistently, rather than setWantedTyBind,
so that unifications are properly tracked.
* Refactoring around solveWantedsTcM, solveWantedsAndDrop.
They previously guaranteed a zonked result, but it's more
straightforward for clients to zonk.
Diffstat (limited to 'testsuite/tests/module')
-rw-r--r-- | testsuite/tests/module/mod71.stderr | 14 |
1 files changed, 3 insertions, 11 deletions
diff --git a/testsuite/tests/module/mod71.stderr b/testsuite/tests/module/mod71.stderr index 12962aa473..53d697e133 100644 --- a/testsuite/tests/module/mod71.stderr +++ b/testsuite/tests/module/mod71.stderr @@ -1,20 +1,12 @@ -mod71.hs:4:9: +mod71.hs:4:9: error: Found hole ‘_’ with type: t1 Where: ‘t1’ is a rigid type variable bound by - the inferred type of f :: (t1 -> a -> t) -> t at mod71.hs:4:1 + the inferred type of f :: Num a => (t1 -> a -> t) -> t + at mod71.hs:4:1 Relevant bindings include x :: t1 -> a -> t (bound at mod71.hs:4:3) f :: (t1 -> a -> t) -> t (bound at mod71.hs:4:1) In the first argument of ‘x’, namely ‘_’ In the expression: x _ 1 In an equation for ‘f’: f x = x _ 1 - -mod71.hs:4:11: - No instance for (Num a) arising from the literal ‘1’ - Possible fix: - add (Num a) to the context of - the inferred type of f :: (t1 -> a -> t) -> t - In the second argument of ‘x’, namely ‘1’ - In the expression: x _ 1 - In an equation for ‘f’: f x = x _ 1 |