Fix #11254.

This moves the call to tcSubType into the context of the
checkInstConstraints call, allowing the deferred type error
somewhere to hang its hat.

2 files changed, 26 insertions, 1 deletions

diff --git a/testsuite/tests/typecheck/should_compile/T11254.stderr b/testsuite/tests/typecheck/should_compile/T11254.stderr
new file mode 100644
index 0000000000..25cd751aa7
--- /dev/null
+++ b/testsuite/tests/typecheck/should_compile/T11254.stderr
@@ -0,0 +1,25 @@
+
+T11254.hs:16:10: warning:
+    • Couldn't match type ‘Frac Int’ with ‘Int’
+        arising from the superclasses of an instance declaration
+    • In the instance declaration for ‘ID Rational’
+
+T11254.hs:16:10: warning:
+    • No instance for (Fractional Int)
+        arising from the superclasses of an instance declaration
+    • In the instance declaration for ‘ID Rational’
+
+T11254.hs:16:10: warning:
+    • No instance for (ID Int)
+        arising from the superclasses of an instance declaration
+    • In the instance declaration for ‘ID Rational’
+
+T11254.hs:18:12: warning:
+    • Couldn't match type ‘GHC.Real.Ratio Integer’ with ‘Int’
+      Expected type: Rational -> Frac Rational
+        Actual type: Rational -> Rational
+    • When checking that instance signature for ‘embed’
+        is more general than its signature in the class
+        Instance sig: Rational -> Rational
+           Class sig: Rational -> Frac Rational
+      In the instance declaration for ‘ID Rational’
diff --git a/testsuite/tests/typecheck/should_compile/all.T b/testsuite/tests/typecheck/should_compile/all.T
index 0c1d0c1d98..bf43716e2a 100644
--- a/testsuite/tests/typecheck/should_compile/all.T
+++ b/testsuite/tests/typecheck/should_compile/all.T
@@ -486,5 +486,5 @@ test('T10935', normal, compile, [''])
 test('T10971a', normal, compile, [''])
 test('T11237', normal, compile, [''])
 test('T10592', normal, compile, [''])
-test('T11254', expect_broken(11254), compile, [''])
 test('T11305', normal, compile, [''])
+test('T11254', normal, compile, [''])