rts: Rip out support for STM invariants

This feature has some very serious correctness issues (#14310),
introduces a great deal of complexity, and hasn't seen wide usage.
Consequently we are removing it, as proposed in Proposal #77 [1]. This
is heavily based on a patch from fryguybob.

Updates stm submodule.

[1] https://github.com/ghc-proposals/ghc-proposals/pull/77

Test Plan: Validate

Reviewers: erikd, simonmar, hvr

Reviewed By: simonmar

Subscribers: rwbarton, thomie, carter

GHC Trac Issues: #14310

Differential Revision: https://phabricator.haskell.org/D4760

4 files changed, 11 insertions, 41 deletions

diff --git a/libraries/base/GHC/Conc.hs b/libraries/base/GHC/Conc.hs
index 8c5c1536d9..15397422a5 100644
--- a/libraries/base/GHC/Conc.hs
+++ b/libraries/base/GHC/Conc.hs
@@ -74,8 +74,6 @@ module GHC.Conc
         , orElse
         , throwSTM
         , catchSTM
-        , alwaysSucceeds
-        , always
         , TVar(..)
         , newTVar
         , newTVarIO
diff --git a/libraries/base/GHC/Conc/Sync.hs b/libraries/base/GHC/Conc/Sync.hs
index 94601f356d..33709d4341 100644
--- a/libraries/base/GHC/Conc/Sync.hs
+++ b/libraries/base/GHC/Conc/Sync.hs
@@ -74,8 +74,6 @@ module GHC.Conc.Sync
         , orElse
         , throwSTM
         , catchSTM
-        , alwaysSucceeds
-        , always
         , TVar(..)
         , newTVar
         , newTVarIO
@@ -777,43 +775,6 @@ catchSTM (STM m) handler = STM $ catchSTM# m handler'
                      Just e' -> unSTM (handler e')
                      Nothing -> raiseIO# e
 
--- Invariant checking has been removed. See #14324 and
--- https://github.com/ghc-proposals/ghc-proposals/blob/master/proposals/0011-deprecate-stm-invariants.rst
-{-# DEPRECATED checkInv, always, alwaysSucceeds
-    [ "The STM invariant-checking mechanism is deprecated in GHC 8.4"
-    , "and will be removed in GHC 8.10. See "
-    , "<https://github.com/ghc-proposals/ghc-proposals/blob/master/proposals/0011-deprecate-stm-invariants.rst>."
-    , ""
-    , "Existing users are encouraged to encapsulate their STM"
-    , "operations in safe abstractions which can perform the invariant"
-    , "checking without help from the runtime system."
-    ] #-}
-
--- | Low-level primitive on which 'always' and 'alwaysSucceeds' are built.
--- 'checkInv' differs from these in that,
---
--- 1. the invariant is not checked when 'checkInv' is called, only at the end of
--- this and subsequent transactions
--- 2. the invariant failure is indicated by raising an exception.
-checkInv :: STM a -> STM ()
-checkInv (STM m) = STM (\s -> case (check# m) s of s' -> (# s', () #))
-
--- | 'alwaysSucceeds' adds a new invariant that must be true when passed
--- to 'alwaysSucceeds', at the end of the current transaction, and at
--- the end of every subsequent transaction.  If it fails at any
--- of those points then the transaction violating it is aborted
--- and the exception raised by the invariant is propagated.
-alwaysSucceeds :: STM a -> STM ()
-alwaysSucceeds i = do ( i >> retry ) `orElse` ( return () )
-                      checkInv i
-
--- | 'always' is a variant of 'alwaysSucceeds' in which the invariant is
--- expressed as an @STM Bool@ action that must return @True@.  Returning
--- @False@ or raising an exception are both treated as invariant failures.
-always :: STM Bool -> STM ()
-always i = alwaysSucceeds ( do v <- i
-                               if (v) then return () else ( errorWithoutStackTrace "Transactional invariant violation" ) )
-
 -- |Shared memory locations that support atomic memory transactions.
 data TVar a = TVar (TVar# RealWorld a)
 
diff --git a/libraries/base/changelog.md b/libraries/base/changelog.md
index 7411aad56e..c588b21487 100644
--- a/libraries/base/changelog.md
+++ b/libraries/base/changelog.md
@@ -3,6 +3,17 @@
 ## 4.12.0.0 *TBA*
   * Bundled with GHC *TBA*
 
+  * The STM invariant-checking mechanism (`always` and `alwaysSucceeds`), which
+    was deprecated in GHC 8.4, has been removed (as proposed in
+    <https://github.com/ghc-proposals/ghc-proposals/blob/master/proposals/0011-deprecate-stm-invariants.rst>).
+    This is a bit earlier than proposed in the deprecation pragma included in
+    GHC 8.4, but due to community feedback we decided to move ahead with the
+    early removal.
+
+    Existing users are encouraged to encapsulate their STM operations in safe
+    abstractions which can perform the invariant checking without help from the
+    runtime system.
+
   * Add a new module `GHC.ResponseFile` (previously defined in the `haddock`
     package). (#13896)
 
diff --git a/libraries/stm b/libraries/stm
-Subproject 33a36c33de150f562a98803e2fc332f07bb2945
+Subproject 8c4d0fabb15ad00beb1e15d027825c78b2c3988