Reviewers: austin, hvr, bgamari

Reviewed By: bgamari

Subscribers: RyanGlScott, dfeuer, adamgundry, rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D4002

After typechecking a data constructor's type signature, its type
variables are partitioned into two distinct groups: the universally
quantified type variables and the existentially quantified type
variables. Then, when prompted for the type of the data constructor,
GHC gives this:

```lang=haskell
MkT :: forall <univs> <exis>. (...)
```

For H98-style datatypes, this is a fine thing to do. But for GADTs,
this can sometimes produce undesired results with respect to
`TypeApplications`. For instance, consider this datatype:

```lang=haskell
data T a where
  MkT :: forall b a. b -> T a
```

Here, the user clearly intended to have `b` be available for visible
type application before `a`. That is, the user would expect
`MkT @Int @Char` to be of type `Int -> T Char`, //not//
`Char -> T Int`. But alas, up until now that was not how GHC
operated—regardless of the order in which the user actually wrote
the tyvars, GHC would give `MkT` the type:

```lang=haskell
MkT :: forall a b. b -> T a
```

Since `a` is universal and `b` is existential. This makes predicting
what order to use for `TypeApplications` quite annoying, as
demonstrated in #11721 and #13848.

This patch cures the problem by tracking more carefully the order in
which a user writes type variables in data constructor type
signatures, either explicitly (with a `forall`) or implicitly
(without a `forall`, in which case the order is inferred). This is
accomplished by adding a new field `dcUserTyVars` to `DataCon`, which
is a subset of `dcUnivTyVars` and `dcExTyVars` that is permuted to
the order in which the user wrote them. For more details, refer to
`Note [DataCon user type variables]` in `DataCon.hs`.

An interesting consequence of this design is that more data
constructors require wrappers. This is because the workers always
expect the first arguments to be the universal tyvars followed by the
existential tyvars, so when the user writes the tyvars in a different
order, a wrapper type is needed to swizzle the tyvars around to match
the order that the worker expects. For more details, refer to
`Note [Data con wrappers and GADT syntax]` in `MkId.hs`.

Test Plan: ./validate

Reviewers: austin, goldfire, bgamari, simonpj

Reviewed By: goldfire, simonpj

Subscribers: ezyang, goldfire, rwbarton, thomie

GHC Trac Issues: #11721, #13848

Differential Revision: https://phabricator.haskell.org/D3687

Tamar Christina authored 7 years ago and Ben Gamari committed 7 years ago

On Windows process creations are fairly expensive. As such calling them in
what's essentially a hot loop is also fairly expensive.

Each time we make a call to `tryGCC` the following fork/exec/wait happen

```
gcc -> realgcc -> cc1
```

This is very problematic, because according to the profiler about 20% of the
time is spent on just process creation and spin time.

The goal of the patch is to mitigate this by asking GCC once for it's search
directories, caching these (because it's very hard to change these at all
after the process started since GCC's base dirs don't change unless with
extra supplied `-B` flags.).

We also do the same for the `findSysDll` function, since this computes
the search path every time by registery accesses etc.

These changes and D3909 drop GHC on Windows startup time from 2-3s to 0.5s.

The remaining issue is a 1.5s wait lock on `CONIN$` which can be addressed
with the new I/O manager code. But this makes GHCi as responsive on Windows as
GHC 7.8 was.

Test Plan: ./validate

Reviewers: austin, hvr, bgamari, erikd

Reviewed By: bgamari

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D3910

Tamar Christina authored 7 years ago and Ben Gamari committed 7 years ago

It's often hard to debug things like segfaults on Windows,
mostly because gdb isn't always of use and users don't know
how to effectively use it.

This patch provides a way to create a crash drump by passing

`+RTS --generate-crash-dumps` as an option. If any unhandled
exception is triggered a dump is made that contains enough
information to be able to diagnose things successfully.

Currently the created dumps are a bit big because I include
all registers, code and threads information.

This looks like

```
$ testsuite/tests/rts/derefnull.run/derefnull.exe +RTS
--generate-crash-dumps

Access violation in generated code when reading 0000000000000000
Crash dump created. Dump written to:
        E:\msys64\tmp\ghc-20170901-220250-11216-16628.dmp
```

Test Plan: ./validate

Reviewers: austin, hvr, bgamari, erikd, simonmar

Reviewed By: bgamari, simonmar

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D3912

This updates the base-4.10.0.0 entry heading which has diverged from

 http://hackage.haskell.org/package/base-4.10.0.0/src/changelog.md

and while at it also sets the GHC version for the base-4.11 entry to
avoid confusion about what GHC 8.2.2's base is going to include.

[skip ci]

This reverts commit 00ff0235.
This reverts commit 11a59de2.

Kind equalities saves the day!

When a record contruction or pattern uses a data constructor
that isn't in scope, we may produce spurious ambiguous-field
errors (Trac #14307).  E.g.

   f (A { fld = x }) = e

where 'A' is not in scope.  We want to draw attention to the
out-of-scope data constructor first; once that is fixed we
can think about the fields.

This patch suppresses the field errors if the data con is out
of scope.

This dark corner was exposed by Trac #14285.  It involves the
interaction between absence analysis and INLINABLE pragmas.

There is a full explanation in Note [aBSENT_ERROR_ID] in MkCore,
which you can read there.  The changes in this patch are

* Make exprIsHNF return True for absentError, treating
  absentError like an honorary data constructor.

* Make absentError /not/ be diverging, unlike other error Ids.

This is all a bit horrible.

* While doing this I found that exprOkForSpeculation didn't
  have a case for value lambdas so I added one.  It's not
  really called on lifted types much, but it seems like the
  right thing

I came across this when debugging something else.  Making it strict
improves the code slightly without affecting behaviour.

Trac #13943 showed that the relatively-new short-cut solver
for class constraints (aka -fsolve-constant-dicts) was wrong.
In particular, see "Type families" under Note [Shortcut solving]
in TcInteract.

The short-cut solver recursively solves sub-goals, but it doesn't
flatten type-family applications, and as a result it erroneously
thought that C (F a) cannot possibly match (C 0), which is
simply untrue.  That led to an inifinte loop in the short-cut
solver.

The significant change is the one line

+                 , all isTyFamFree preds  -- See "Type families" in
+                                          -- Note [Shortcut solving]

but, as ever, I do some other refactoring.  (E.g. I changed the
name of the function to shortCutSolver rather than the more
generic trySolveFromInstance.)

I also made the short-cut solver respect the solver-depth limit,
so that if this happens again it won't just produce an infinite
loop.

A bit of other refactoring, notably moving isTyFamFree
from TcValidity to TcType

Summary:
As reported by Alex Lang, R_X86_64_NONE relocations
appear in per-package object files, not per-module object
files. This diff adds _NONE relocations for x86.

Reviewers: bgamari, geekosaur, austin, erikd, simonmar

Reviewed By: geekosaur

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D4062

Summary:
building libffi docs with our intree-libffi seems
rather pointless.

Reviewers: bgamari, austin

Reviewed By: bgamari

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D4054

[skip ci]

[skip ci]

Ben Gamari authored 7 years ago and Ben Gamari committed 7 years ago

One step closer to being able to drop the Windows Perl tarball. We
previously attempted to do this in D3567 but were forced to revert due
to Windows problems.

Acknowledgements:
 * @Phyx kindly contributed the codepath allowing this to work on
Windows.

Test Plan: Validate

Reviewers: hvr, austin, Phyx

Subscribers: erikd, thomie, rwbarton

Differential Revision: https://phabricator.haskell.org/D3918

    class (a `C` b) c

Is pretty printed as

    class a `C` b c

Fixes #14306

Commit 063e0b4e accidentially undid
7b8827ab where I bumped nofib
previously.

Moritz Angermann authored 7 years ago and Ben Gamari committed 7 years ago

This is rather annoying. I'd prefer to have a stable release to
use. However libffi-3.2.1 has been released November 12, 2014, and
libffi-4 is TBD. See also https://github.com/libffi/libffi/issues/296

The core reason for this change is that llvm changed the supported
assembly to unified syntax, which libffi-3.2.1 does not use, and hence
fails to compile for arm with llvm. For refence, see the following
issue: https://github.com/libffi/libffi/issues/191.

This diff contains a script to generate a tarball for the
`libffi-tarballs` repository from the libffi GitHub repository; as well
as the necessary changes to the build system.

Updates libffi-tarballs submodule.

Reviewers: austin, bgamari, hvr

Subscribers: hvr, erikd, rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D3574

Ben Gamari authored 7 years ago and Ben Gamari committed 7 years ago

Test Plan: Validate

Reviewers: austin, snowleopard

Subscribers: angerman, rwbarton, thomie, int-e

Differential Revision: https://phabricator.haskell.org/D4042

Ben Gamari authored 7 years ago and Ben Gamari committed 7 years ago

Character is a terribly overloaded term and may refer to graphemes or
code points. Specifically say that Char represents Unicode code points.

[skip ci]

Test Plan: Read it.

Reviewers: hvr, austin

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D4051

Inspired by the discussion in #14296, I've decided to
document a law which is usually in the back of my mind when I'm using
Template Haskell's `Lift` class, but isn't formally stated anywhere.
That is, every `Lift` instance should satisfy (for all `x`):

```lang=haskell
$(lift x) == x
```

Test Plan: Read it

Reviewers: austin, goldfire, bgamari

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D4050

Ben Gamari authored 7 years ago and Ben Gamari committed 7 years ago

[skip ci]

Reviewers: hvr, austin, dfeuer

Reviewed By: dfeuer

Subscribers: rwbarton, thomie

GHC Trac Issues: #14284

Differential Revision: https://phabricator.haskell.org/D4043

Summary:
For some reson, the `*`, that was supposed to be in D4031
got lost in the diff.

Reviewers: bgamari, austin, hvr, dfeuer

Reviewed By: dfeuer

Subscribers: rwbarton, thomie, erikd

Differential Revision: https://phabricator.haskell.org/D4044

The check is a bit gnarly, but I couldn't think of a better way.
See the new code in TcTyClsDecls.

test case: polykinds/T13391

Simon Marlow authored 7 years ago and Ben Gamari committed 7 years ago

The strictness signature for a data con wrapper wasn't including any
dictionary arguments, which meant that bangs on the fields of a
constructor with an existential context would be moved to the wrong
fields.  See T14290 for an example.

Test Plan:
* New test T14290
* validate

Reviewers: simonpj, niteria, austin, bgamari, erikd

Reviewed By: simonpj, bgamari

Subscribers: rwbarton, thomie

GHC Trac Issues: #14290

Differential Revision: https://phabricator.haskell.org/D4040

Ben Gamari authored 7 years ago and Ben Gamari committed 7 years ago

Previously if we had both ld.lld and ld.gold installed but a gcc which
didn't support -fuse-ld=lld we would fail when trying ld.lld and fall
immediately back to plain ld. Now we will try ld.gold as well. This was
brought to light by #14280, where using ld.bfd resulted in a broken
stage2 compiler.

Test Plan: Configure

Reviewers: angerman, hvr, austin

Reviewed By: angerman

Subscribers: rwbarton, thomie, erikd

GHC Trac Issues: #14280

Differential Revision: https://phabricator.haskell.org/D4038

Niklas Hambüchen authored 7 years ago and Ben Gamari committed 7 years ago

Reviewers: bgamari, austin, hvr, Phyx

Reviewed By: Phyx

Subscribers: Phyx, rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D4010

Niklas Hambüchen authored 7 years ago and Ben Gamari committed 7 years ago

Btw, -Wconversion is off by default and not included in -Wall, -Wextra
or -pedantic, so I used it temporarily with -optc-Wconversion.

Reviewers: bgamari, austin, hvr

Reviewed By: bgamari

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D3965

Niklas Hambüchen authored 7 years ago and Ben Gamari committed 7 years ago

Knowing this is important for followup commits, where we will subtract
getProcessElapsedTime() values from each other, in a way that assumes
that there is no wrapping every 49 days.

Reviewers: bgamari, austin, erikd, simonmar, NicolasT

Reviewed By: bgamari

Subscribers: rwbarton, thomie

GHC Trac Issues: #14233

Differential Revision: https://phabricator.haskell.org/D3964

Ben Gamari authored 7 years ago and Ben Gamari committed 7 years ago

These are needed by the testsuite and consequently must be shipped in
the testsuite tarball to ensure that we can test binary distributions.

See #13897.

Test Plan: Validate

Reviewers: austin

Subscribers: snowleopard, rwbarton, thomie

GHC Trac Issues: #13897

Differential Revision: https://phabricator.haskell.org/D4039

Summary:
The llvm-targets file records `mattr` values, and
while interrogating `clang` for the target, we might
stumble upon `+soft-float-abi`, however ghc does not support
full soft-float, and as such passing `+soft-float` to `llc`
will result in segfaults for any function passing float
registers F1, ... in the ARM Instruction Selection Pass.

Reviewers: bgamari, austin

Reviewed By: bgamari

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D4030

Summary:
Our usual GHC_CONVERT_OS macro, will turn any andoird* into android.
This however drops the essential androideabi part. As such for the
GHC_LLVM_TARGET we only convert the VENDOR, not the OS.

Reviewers: bgamari, austin, hvr

Reviewed By: bgamari

Subscribers: rwbarton, thomie, erikd

Differential Revision: https://phabricator.haskell.org/D4031

Tamar Christina authored 7 years ago and Ben Gamari committed 7 years ago

Exception handling on Windows is unfortunately a bit complicated.
But essentially the VEH Handlers we currently have are running too
early.

This was a problem as it ran so early it also swallowed C++ exceptions
and other software exceptions which the system could have very well
recovered from.

So instead we use a sequence of chains to for the exception handlers to
run as late as possible. You really can't get any later than this.

Please read the comment in the patch for more details.

I'm also providing a switch to allow people to turn off the exception
handling entirely. In case it does present a problem with their code.

(Reverted and recommitted to fix authorship information)

Test Plan: ./validate

Reviewers: austin, hvr, bgamari, erikd, simonmar

Reviewed By: bgamari

Subscribers: rwbarton, thomie

GHC Trac Issues: #13911, #12110

Differential Revision: https://phabricator.haskell.org/D3911

Reverting to fix authorship of commit.

This reverts commit 1825cbdb.