Andrew Martin authored 5 years ago and Marge Bot committed 5 years ago

GHC Proposal: 0013-unlifted-newtypes.rst
Discussion: https://github.com/ghc-proposals/ghc-proposals/pull/98


Issues: #15219, #1311, #13595, #15883
Implementation Details:
  Note [Implementation of UnliftedNewtypes]
  Note [Unifying data family kinds]
  Note [Compulsory newtype unfolding]

This patch introduces the -XUnliftedNewtypes extension. When this
extension is enabled, GHC drops the restriction that the field in
a newtype must be of kind (TYPE 'LiftedRep). This allows types
like Int# and ByteArray# to be used in a newtype. Additionally,
coerce is made levity-polymorphic so that it can be used with
newtypes over unlifted types.

The bulk of the changes are in TcTyClsDecls.hs. With -XUnliftedNewtypes,
getInitialKind is more liberal, introducing a unification variable to
return the kind (TYPE r0) rather than just returning (TYPE 'LiftedRep).
When kind-checking a data constructor with kcConDecl, we attempt to
unify the kind of a newtype with the kind of its field's type. When
typechecking a data declaration with tcTyClDecl, we again perform a
unification. See the implementation note for more on this.

Co-authored-by: Richard Eisenberg <rae@richarde.dev>

Alp Mestanogullari authored 5 years ago and Marge Bot committed 5 years ago

Each package's object files were 'need'ing the library files of all transitive
dependencies of the current package, whichi is pointless since the said
libraries are not needed until we link those object files together.

This fixes #16759.

Ömer Sinan Ağacan authored 5 years ago and Marge Bot committed 5 years ago

Fixes #16696

Ben Gamari authored 5 years ago and Marge Bot committed 5 years ago

Previously we would pass flags intended for the C compiler to the C++
compiler (see #16738). This would cause, for instance, `-std=gnu99` to
be passed to the C++ compiler, causing spurious test failures. Fix this
by maintaining a separate set of flags for C++ compilation invocations.

Ben Gamari authored 5 years ago and Marge Bot committed 5 years ago

Enable artifacts on to ensure we have bindist coverage.

Ben Gamari authored 5 years ago and Marge Bot committed 5 years ago

This should help alleviate queue times as the LLVM job is one of the
longest that we have.

Ben Gamari authored 5 years ago and Marge Bot committed 5 years ago

Previously we used an awful hybrid batch script/Bourne shell script to
allow this test to run both on Windows and Linux (fixing #9399).
However, this breaks on some libc implementations (e.g. musl). Fix this.

Fixes #16798.

Ben Gamari authored 5 years ago and Marge Bot committed 5 years ago

Move prepare-system.sh to .gitlab and remove everything else.

Ben Gamari authored 5 years ago and Marge Bot committed 5 years ago

For reasons I don't understand sphinx seems to fail to produce a .idx
file for makeindex.

Ben Gamari authored 5 years ago and Marge Bot committed 5 years ago

Strangely these were only causing the build to fail in the aarch64-linux
job, despite Sphinx throwing errors in all jobs I checked.

Also changes some `#ifdef`s to `#if defined` to satisfy the linter.

Alp Mestanogullari authored 5 years ago and Marge Bot committed 5 years ago

MR !1151 makes the Hadrian/Linux job pass by fixing the last two test failures,
so we can now be stricter and not allow those jobs to fail anymore, easily
letting us see when patches introduce test failures.

davide authored 5 years ago and Marge Bot committed 5 years ago

This requires creating RTS library symlinks when registering, outside
of the rule for the registered library file.

Ben Gamari authored 5 years ago and Marge Bot committed 5 years ago

Previously the presence of source notes could hide nested applications
of `unpackFoldrCString#` from our constant folding logic. For instance,
consider the expression:

```haskell
unpackFoldrCString# "foo" c (unpackFoldrCString# "baz" c n)
```

Specifically, ticks appearing in two places can defeat the rule:

  a. Surrounding the inner application of `unpackFoldrCString#`
  b. Surrounding the fold function, `c`

The latter caused the `str_rules` testcase to fail when `base` was built
with `-g3`.

Fixes #16740.

Phuong Trinh authored 5 years ago and Marge Bot committed 5 years ago

`checkUnload` currently doesn't check the info header of static objects.
Thus, it may free an `ObjectCode` struct wrongly even if there's still a
live static object whose info header lies in a mapped section of that
`ObjectCode`. This fixes the issue by adding an appropriate check.

Ben Gamari authored 5 years ago and Marge Bot committed 5 years ago

This should finally fix #14261.

[skip ci]

Fixes linters.

Since we can't load profiled objects when GhcDynamic==YES. Affects:

* T16737
* T16384
* T16718
* T16619
* T16190

galois_raytrace was previously broken on i386 due to use of x87
arithmethic on that platform. However,
42504f4a removes x87 support; this
resulted in an unexpected pass. Unmark this test as broken.

This test requires FFI usage.

 * Skip process005 in ghci way
 * Mark process002 as fragile in threaded2

Marks posix002 as fragile in threaded2 way due to #16550.

The linter now enforces our preference for `#if defined()` and
`#if !defined()`.

Previously we would apply this rule, which is only intended for
testsuite Makefiles, to all files. This lead to a number of
false-positives in all.T files.

I earlier accidentally corrected it breaking the test.

T493 tests #493, which is an FFI test. FFI tests should be skipped
in ghci way.

cc017 requires TH but we can't load dynamic profiled objects.

As noted in #16555.

omit_ways expects a list but this was broken in several cases.

It times out pretty reliably. It's not clear that much is gained by
running this test in the ghci way anyways.

It was previously marked as broken but it passes non-deterministically.
See #2783.

As noted in #16543.

As noted in #16542, the expected rule doesn't fire. However, this
doesn't seem terribly surpring given the circumstances.

As noted in #16541.

As noted in #16540.

Due to #16537.