GitLab

The script I used is included as testsuite/driver/kill_extra_files.py,
though at this point it is for mostly historical interest.

Some of the tests in libraries/hpc relied on extra_files.py, so this
commit includes an update to that submodule.

One test in libraries/process also relies on extra_files.py, but we
cannot update that submodule so easily, so for now we special-case it
in the test driver.

Some of the *.T files were in libraries/hpc, so this contains an
update to that submodule.

Summary: My script won't understand this.

Test Plan: validate

Reviewers: austin, bgamari, simonmar

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3210

This commit fixes several bugs related to case expressions
involving numeric literals which are not in the range of values of
their (fixed-width, integral) type.

There is a new invariant on Literal: The argument of a MachInt[64]
or MachWord[64] must lie within the range of the corresponding
primitive type Int[64]# or Word[64]#, as defined by the target machine.
This invariant is enforced in mkMachInt[64]/mkMachWord[64] by wrapping
the argument to the target type's range if necessary.

Test Plan: Test Plan: make slowtest TEST="T9533 T9533b T9533c T10245
T10246"

Trac issues: #9533, #10245, #10246, #13171

Reviewers: simonmar, simonpj, austin, bgamari, nomeata

Reviewed By: bgamari

Subscribers: thomie, rwbarton

Differential Revision: https://phabricator.haskell.org/D810

Trac issues: #13194

Reviewers: austin, hvr, erikd, bgamari, dfeuer, duncan

Subscribers: DemiMarie, dfeuer, thomie

Differential Revision: https://phabricator.haskell.org/D3090

Reviewers: austin

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3184

Bumps binary submodule.

While working through the FunCo patch I encountered some lint issues
which suggested that `Constraint` wasn't being considered equal to `TYPE
'LiftedRep`.  Consequently I introduced this axiom and associated messy
ball of logic to explicitly coerce `Constraint`.

However, as @goldfire points out on D3208 this really shouldn't be
necessary.  Indeed, I tried ripping out the axiom and things appear to
just work. I suspect the issue motivating the axiom was a bug elsewhere
in the FunCo branch that I fixed during development.

Test Plan: Validate

Reviewers: simonpj, goldfire, austin

Reviewed By: goldfire

Subscribers: thomie, goldfire

Differential Revision: https://phabricator.haskell.org/D3218

The build system uses the CrossCompiling variable to decide whether or
not we should build various packages that must be built using the
compiler.  Consequently, it is important that we persist its value in
the binary distribution so we know during `make install` not to go
looking for files that would have been built for these packages. Failing
to do this causes #13325.

Test Plan: Cross compile, `make binary-dist`, and try installing the
binary distribution on the target

Reviewers: hvr, austin, trofi, rwbarton

Reviewed By: trofi, rwbarton

Subscribers: carter, trofi, rwbarton, erikd, thomie, snowleopard, davean

Differential Revision: https://phabricator.haskell.org/D3187

This too is prepartory for my early-inlining patch.  It turned
out that early inlining exposed a bug in the way that static
pointers were being floated.

Reviewers: austin, bgamari

Reviewed By: bgamari

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3205

Reviewers: austin, bgamari

Reviewed By: bgamari

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3200

Test Plan: n/a

Reviewers: bgamari, austin

Reviewed By: bgamari

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3144

Test Plan: Run test of `openTempFile` under `strace` to verify
reasonably random filenames

Reviewers: austin, hvr, trofi, rwbarton

Reviewed By: trofi, rwbarton

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D3188

Test Plan: Validate

Reviewers: rwbarton, austin

Reviewed By: rwbarton

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3215

Although I'm really not convinced that measuring peak_megabytes_allocated of
this test is reasonable.

The GCC Bindists that we use compile with `pthread` enabled by default.
This means that on every link the dll is passed as a dependency by the
driver. Lots of packages depend on it but the runtime linker doesn't
provide it by default making compiled code work but not interpreted.

Following D3028 `pthreads` would be provided by default ONLY when linked
dynamicly, which we don't support yet (See D2592). Until this is the
case we need to manually provide `libpthreads`.

Test Plan: ./validate

Reviewers: austin, hvr, bgamari

Reviewed By: bgamari

Subscribers: thomie, #ghc_windows_task_force

Differential Revision: https://phabricator.haskell.org/D3155

We have a unfortunate workaround in place for the fact that most
packages out there use POSIX names for symbols even on Windows.  This
means that we have to recognize e.g. both `_ungetch` and `ungetch`.

The former is the actual symbol name on windows and the latter is the
POSIX variant. The problem is that on normal windows programs `ungetch`
should not be in the global namespace.

To work around this, we now mark the deprecated symbols as weak symbols
in the global namespace. This provides the flexibility we need:

* If you use the symbol without defining it, we assume you meant to use
  the POSIX variant.
* If you actually define the symbol, we'll hence forth use that
  definition and assume you didn't mean to use POSIX code. This is how
  MingW64's wrapper also works.

This requires D3028.

Fixes #13210.

Test Plan: ./validate

Reviewers: austin, bgamari, erikd, simonmar

Reviewed By: bgamari

Subscribers: thomie, #ghc_windows_task_force

Differential Revision: https://phabricator.haskell.org/D3154

When the `GCC` driver envokes the pipeline a `SPEC` is used to determine
how to configure the compiler and which libraries to pass along.

For Windows/mingw, this specfile is
https://github.com/gcc-mirror/gcc/blob/master/gcc/config/i386/mingw32.h

This has a lot of interesting things that we need to emulate in order to
be able to link as many things out of the box as GCC. In particular this
is why you never need to specify `-lgcc_s` when compiling, but you do
when using `GHCi`.

Unfortunately due to time constraints I can't set up the framework
required in `GHC` to do this before the feature freeze.

So I suggest this alternate implementation:
When we load a dll, also bring it's dependencies into scope of the
interpeter.

This has pros and cons. Pro is, we'll fix many packages on hackage which
specify just `-lstdc++`. Since this points to `libstdc++-6.dll` which
will bring `libgcc` into scope.

The downside is, we'll be more lenient than GCC, in that the interpreter
will link much easier since it has implicit dependencies in scope.
Whereas for compilation to work you will have to specify it as an
argument to GHC.

This will make the Windows runtime linker more consistent with the unix
ones. The difference in semantics came about because of the differences
between `dlsym` and `GetProcAddress`.  The former seems to search the
given library and all it's dependencies, while the latter only searches
the export table of the library. So we need some extra manual work to
search the dependencies which this patch provides.

Test Plan:
```
./validate
```
```
$ echo :q | inplace/bin/ghc-stage2.exe --interactive +RTS -Dl -RTS
-lstdc++ 2>&1 | grep "Loading dependency"
```

```
$ echo :q | ../inplace/bin/ghc-stage2.exe --interactive -lstdc++ +RTS
-Dl -RTS 2>&1 | grep "Loading dependency"
Loading dependency *.exe -> GDI32.dll.
Loading dependency GDI32.dll -> ntdll.dll.
Loading dependency *.exe -> KERNEL32.dll.
Loading dependency KERNEL32.dll -> KERNELBASE.dll.
Loading dependency *.exe -> msvcrt.dll.
Loading dependency *.exe -> SHELL32.dll.
Loading dependency SHELL32.dll -> USER32.dll.
Loading dependency USER32.dll -> win32u.dll.
Loading dependency *.exe -> WINMM.dll.
Loading dependency WINMM.dll -> WINMMBASE.dll.
Loading dependency *.exe -> WSOCK32.dll.
Loading dependency WSOCK32.dll -> WS2_32.dll.
Loading dependency WS2_32.dll -> RPCRT4.dll.
Loading dependency libstdc++-6.dll -> libwinpthread-1.dll.
Loading dependency libstdc++-6.dll -> libgcc_s_seh-1.dll.
```

Trac tickets: #13093, #13189

Reviewers: simonmar, rwbarton, austin, bgamari, erikd

Reviewed By: bgamari

Subscribers: rwbarton, RyanGlScott, thomie, #ghc_windows_task_force

Differential Revision: https://phabricator.haskell.org/D3028

Summary:
The plan is to release a separate library, 'compact', which gives a
friendly user-facing interface.  This library is just enough so that we
can make sure the functionality is working in GHC.
Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Test Plan: validate

Reviewers: bgamari, dfeuer, austin, simonmar, hvr

Subscribers: thomie, erikd, snowleopard

Differential Revision: https://phabricator.haskell.org/D3206

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Summary:
The previous implementation tried to be "efficient" by
storing field names once in IfaceConDecls, and only just
enough information for us to reconstruct the FieldLabel.
But this came at a bit of code complexity cost.

This patch undos the optimization, instead storing a full
FieldLabel at each data constructor.  Consequently, this fixes
bugs #12699 and #13250.
Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Test Plan: validate

Reviewers: adamgundry, bgamari, austin

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3174

`Applicative` as a superclass of `Monad` is non-standard.

Fixes #13196.

[skip ci]

Reviewers: austin, bgamari

Reviewed By: bgamari

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3185

[skip ci]

There are two things here

* Use exprIsLiteral_maybe to "look through" a variable bound
  to a literal string.

* Add CONLIKE to the NOINLINE pragma for unpackCString# and
  unpackCStringUtf8#

See Trac #13317, Trac #10844, and
Note [exprIsConApp_maybe on literal strings] in CoreSubst

I did a nofib run and got essentially zero change except for one
2.2% improvement in allocation for 'pretty'.

`(:~~:)`, the hetergeneous version of `(:~:)`, should have class
instances similar to those of `(:~:)`, especially since their
implementations aren't particularly tricky or surprising. This adds
them.

Reviewers: bgamari, austin, hvr, goldfire

Reviewed By: bgamari

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3181

GHC and ghc-pkg make some pretty hard assumptions about where they're
running on Windows. They assume that they are always running from
`foo/bin/ghc.exe` and that to find the `lib` folder they can drop
`bin/ghc.exe` from the base path and append `lib`.

This is already false for the testsuite, which when testing thenbindist
 has one test which puts the binaries in `inplace/test   spaces`.

For some reason before this was either being skipped or mysteriously 
passing.
But as of `2017.02.11` our luck ran out.

the testsuite triggers a failure such as those in #13310

Let's soften the assumption and just check that `../lib` exists instead.

80 chars

Test Plan: ./validate

Reviewers: austin, erikd, bgamari

Reviewed By: bgamari

Subscribers: thomie, #ghc_windows_task_force

Differential Revision: https://phabricator.haskell.org/D3158

Parse `foo, bar, baz` into `And [foo, bar, baz]`
instead of `And [foo, And [bar, baz]]`.

Fixes #11024.

Test Plan: read and think

Reviewers: austin, bgamari, mpickering

Reviewed By: bgamari, mpickering

Subscribers: ezyang, thomie, alanz

Differential Revision: https://phabricator.haskell.org/D3139

Summary:
Fixes #13323.
Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Test Plan: validate

Reviewers: bgamari, austin

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3175

Reviewers: austin, bgamari

Reviewed By: bgamari

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3179

I don't see any reason for the (filesize*2), and experimentally
allocations do go down slightly after this change.

Test Plan: validate

Reviewers: simonmar, austin, bgamari, trofi

Reviewed By: bgamari, trofi

Subscribers: trofi, thomie

Differential Revision: https://phabricator.haskell.org/D3164

I discovered this feature via Stack Overflow.
https://stackoverflow.com/questions/42310399

Reviewers: austin, bgamari

Reviewed By: bgamari

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D3150

getArgs didn't match the treatmeant of -- in the RTS leading to
inconsistencies between behavior on Windows and other platforms. See #13287.

Reviewers: austin, hvr, bgamari, erikd, simonmar, rwbarton

Reviewed By: bgamari, rwbarton

Subscribers: rwbarton, thomie

Differential Revision: https://phabricator.haskell.org/D3147