GitLab

* Reorganise GHC and GHC.Packages modules to get rid of import cycles
* Move "GHC.Packages" to "Packages", merge "GHC" into "Settings.Default"

#531 removed install rules (`src/Rules/Install.hs`), but left a lot of supporting code that is now unused.

Hopefully, this clean up completes the removal of install rules. If they are ever coming back to Hadrian in any shape, some useful code can be recovered here.

See #540.

* Keep only Stage0 and Stage1 package databases

* Fix default top-level targets

* Drop ghcCabal package
* Merge GhcCabal and CabalFlags builders
* Use more consistent naming
* Resolve import cycle
* Rename parseCabal to parseCabalFile
* Remove outdated references to ghc-cabal from comments

This fixes #654.

There are only two important changes.

1) The first one fixes missing dependencies:

```diff
-pkgDependencies = fmap (fmap PD.dependencies) . readPackageDataFile
+pkgDependencies = fmap (fmap (map pkgName . packageDependencies)) . readCabalFile
```

Here `PD.dependencies` returned versioned package names, e.g. `ghc-8.7`, which then failed to match with non-versioned package names such as `ghc` in `contextDependencies`. Switching from `PD.dependencies` to `packageDependencies` fixes this.

2) I clearly remember that we didn't have this bug before, so I added some tests for our package dependency infrastructure to prevent such regressions in future:

```haskell
testDependencies :: Action ()
testDependencies = do
    putBuild "==== pkgDependencies"
    depLists <- mapM (pkgDependencies . vanillaContext Stage1) ghcPackages
    test $ and [ deps == sort deps | Just deps <- depLists ]
    putBuild "==== Dependencies of the 'ghc-bin' binary"
    ghcDeps <- pkgDependencies (vanillaContext Stage1 ghc)
    test $ isJust ghcDeps
    test $ pkgName compiler `elem` fromJust ghcDeps
    stage0Deps <- contextDependencies (vanillaContext Stage0 ghc)
    stage1Deps <- contextDependencies (vanillaContext Stage1 ghc)
    stage2Deps <- contextDependencies (vanillaContext Stage2 ghc)
    test $ vanillaContext Stage0 compiler `notElem` stage1Deps
    test $ vanillaContext Stage1 compiler `elem`    stage1Deps
    test $ vanillaContext Stage2 compiler `notElem` stage1Deps
    test $ stage1Deps /= stage0Deps
    test $ stage1Deps == stage2Deps
```

Everything else are cosmetic changes, fixing minor issues in comments, and adding TODOs. To figure out the failure in #654 I had to read some code I didn't write and my hands were automatically fixing some style inconsistencies with the rest of the Hadrian code base. (I'd like to emphasise that I make no judgement about which style is better, it's merely an attempt to make the code base look more homogeneous, which I think is useful.)

The fix has landed in GHC HEAD.

Closes #634.

* implement the equivalent of D4991 for hadrian

* use stage2 instead of stage == Stage2

* use @quasicomputational's Cabal branch in CI for the time being

* workaround can't-use-Natural-in-base

* adding include directories

* Adding includes

* Stable cabal sandbox build

* adding include directories

* Adding includes

* Shipping ghci

* Improved GhcCabal library arguments

* setting enviornment for extra programs

* Copied necessary programs

* Added support to generate ghcconfig
  We will need it for properly configuring python command

Some revisions

* Using ghcconfig file for test parameters

* minor changes

* minor revision

* Update Ghc.hs

* Added featues to binary distribution

* Minor revision

* Using takeFileName instead of basename

* Update BinaryDist.hs

* For executables, we should read the `main-is` field from the cabal file.

Previously, we simply treat file name for `Main` module as `Main.hs` to
build executable. That doesn't work for the `timeout` program. This patch
fixes the problem.

* Add comments about the processing of `main-is` field from .cabal file.

* Added option to specify config file

* Added speed setting for validation

* Revision

* attempt at setting some more test arguments correctly

* use vanillaContext

* Minor clean up

* Track rts.cabal

* Move all package-specific settings to Settings.Packages, plus another revision, see #540

* Drop Rules.PackageData

* Rule for testsuite dependencies

* Added RunTest config options

* added support to set test speed with runtest

* Added support for more testing features

* Rectified Merge Errors

* using all available threads

* Minor Revision

* Removed TestThread argument

* Update Utilities.hs

* Added support to choose test compiler

* Minor Revision

* Added comments

* Update RunTest.hs

* Update CommandLine.hs

* Update RunTest.hs

See #540 

See #615

The shake-0.16 can't be built with ghc-8.4.3, due to the
Semigroup-Monoid-Proposal changes. shake-0.16.1 fixed that.

* Rule for testsuite dependencies

* Separated validate builder arguments

* Added RunTest config options

* added support to set test speed with runtest

* Fixed minor bug with testConfigs

Removed indentation error

* Added support for more testing features

* Rectified Merge Errors

* Removed need rule for Hp2ps

* using all available threads

* Minor Revision

* Removed TestThread argument

* Update Utilities.hs

* Update RTS flags

See #611

* More tweaks

See https://phabricator.haskell.org/D4436

* first draft of a nofib rule

* address some of Andrey's feedback

* refactor nofib into a proper Builder, now runs but one of the programs fails

* more subtle error handling, docs

* get rid of RunNofib builder, invoke commands directly

For details, see https://phabricator.haskell.org/D4695

See #593

A nicer fix would involve patching GHC to not just look under
$libexec/ but also under the directory where the GHC binary itself lives
(bin/ for hadrian), so that we can leave all binaries under bin/.

Addresses Trac #15132.