GitLab

The later caused issues on windows by being translated into
"\\dev\\null" and python then trying to open this non-existant file.

So we now use os.devnull inside python and convert it to "/dev/null"
when calling out to the shell, which is bound to run in a unix like
environment.

This fixes an issue a test producing unexpected stderr output failed
with a framework failure instead of showing a diff of the output.

We used to serialise large integers as strings. Now they are serialized
as a list of Bytes.

This changes the size for a Integer in the higher 64bit range from 77 to
9 bytes when written to disk.

The impact on the general case is small (<1% for interface files) as we
don't use many Integers. But for code that uses many this should be a
nice benefit.

The `UniqDFM` is deterministic, of course, while we provide an unsafe
`NonDetUniqFM` wrapper for `UniqFM` to opt into nondeterministic instances.

This does not make any changes in the contents -- formatting only.

Previously the comments were too noisy and I've always found it very
hard to read. Hopefully it's easier to read now.

AndreasK recently mentioned that he thought that interface file loading
may be a non-trivial cost. Let's measure.

`TablesNextToCode` is now a substituted by configure, where it has the
correct defaults and error handling. Nowhere else needs to duplicate
that, though we may want the compiler to to guard against bogus settings
files.

I renamed it from `GhcEnableTablesNextToCode` to `TablesNextToCode` to:

 - Help me guard against any unfixed usages

 - Remove any lingering connotation that this flag needs to be combined
   with `GhcUnreigsterised`.

Original reviewers:

Original subscribers: TerrorJack, rwbarton, carter

Original Differential Revision: https://phabricator.haskell.org/D5082

This reverts commit 4e1dfc37.

Due to #16943.

Supply branch incomps when building an IfaceClosedSynFamilyTyCon

`pprTyThing` now has access to incomps. This also causes them to be
written out to .hi files, but that doesn't pose an issue other than a
more faithful bijection between `tyThingToIfaceDecl` and `tcIfaceDecl`.

The machinery for displaying axiom incomps was already present but not
in use. Since this is now a thing that pops up in ghci's :info the
format was modified to look like a haskell comment.

Documentation and a test for the new feature included.

Test Plan: T15546

Reviewers: simonpj, bgamari, goldfire

Reviewed By: simonpj

Subscribers: rwbarton, carter

GHC Trac Issues: #15546

Differential Revision: https://phabricator.haskell.org/D5097

We introduce a PlatformWordSize type and use it in platformWordSize
field.

This removes to panic/error calls called when platform word size is not
32 or 64. We now check for this when reading the platform config.

I was attempting to load hadrian into ghci by using
`cabal new-repl exe:hadrian` but it failed because it tried
to use this `.ghci` configuration.

I'm not sure who used this script but you should really use the new-repl
method.

Shake will be moving from its current implementation of ?== to one from
System.FilePattern. Support for `//` is being dropped, leaving only `*`
and `**` as special forms. This commit converts the existing file
patterns in Hadrian to the new format. It also removes all occurances
of <//> and changes the user-settings docs to remove references to //
and add **.

The conversion is as follows:

- //a ==> **/a

- a// ==> a/**

- a//b ==> a/**/b

Breaks fragile tests into two groups, allowing us to easily preserve
stdout/stderr of failing fragile tests.

Fedora uses the naming llc-7.0 while Debian uses llc-7. Ensure that both
are found.

Fixes #16990.

Ticket #16978 showed that we were floating a recursive,
unlifted join point to top level.  It's very much a corner
case:

    joinrec j :: Int#
            j = jump j
    in ...

But somehow it showed up in a real program.

For non-recursive bindings in SetLevels.lvlBind we were already
(correctly) checking for unlifted bindings, but when I wrote
that code I didn't think that a /recursive/ binding could be
unlifted but /join-points/ can be!

Actually I don't think that SetLevels should be floating
join points at all.  SetLevels really floats things to move
stuff out of loops and save allocation; but none of that applies
to join points.  The only reason to float join points is in
cases like
   join j1 x = join j2 y = ...
               in ...
which we might want to swizzle to
   join j2 x y = ... in
   join j1 x = ...
   in ...
because now j1 looks small and might be inlined away altogether.
But this is a very local float perhaps better done in the simplifier.

Still: this patch fixes the crash, and does so in a way that is
harmless if/when we change our strategy for floating join points.

Add some files generated by hadrian and some tooling files

For some reason configure seems unable to find it on its own. Let's try
giving it a hint.

Addresses #16398.

This fixes #17003, where a user directly asked for the 'docs-haddock' target
without building a complete stage 2 GHC first. Since haddock only depends on
lib:ghc, the stage 2 GHC executable wasn't built, and neither were the
settings, platformConstants, llvm-passes and llvm-targets files, since they
are declared to be dependencies of exe:ghc.

This makes sense in general since all GHC API users (haddock is one) will likely
want those files to be there.

Previously there were numerous places in the RTS where we would fopen
with the "w" flag string. This is wrong as it will not truncate the
file. Consequently if we write less data than the previous length of the
file we will leave garbage at its end.

Fixes #16993.

This file refers to shell.nix, which was removed in
430e6fed and c00d2f59.

Previously, we detected dodgy type family instances binders by
expanding type synonyms (via `exactTyCoVarsOfType`) and looking for
type variables on the RHS that weren't mentioned on the (expanded)
LHS. But this doesn't account for type families (like the example
in #17008), so we instead use `injectiveVarsOfType` to only count
LHS type variables that are in injective positions. That way, the `a`
in `type instance F (x :: T a) = a` will not count if `T` is a type
synonym _or_ a type family.

Along the way, I moved `exactTyCoVarsOfType` to `TyCoFVs` to live
alongside its sibling functions that also compute free variables.

Fixes #17008.

GHC had an ad hoc validity check in place to rule out pattern
variables bound by type synonyms, such as in the following example:

```hs
type ItemColID a b = Int  -- Discards a,b

get :: ItemColID a b -> ItemColID a b
get (x :: ItemColID a b) = x :: ItemColID a b
```

This hack is wholly unnecessary nowadays, since OutsideIn(X) is more
than capable of instantiating `a` and `b` to `Any`. In light of this,
let's rip out this validity check.

Fixes #17007.

[skip-ci]

1. Slightly increased the waiting time for the tested effect to be more
   profound.

2. Introduced measuring of the actual time spent waiting and adjusing
   CPU time by it to compensate for threadDelay waiting time
   inconsistencies.

It seems that the regular expression parser changed in GitLab 12.1 and
now does now support forward slashes in the RE, even when escaped.

As mentioned in #16997, GHC currently complains about this import.
In general I'm reluctant to paper over things like this but in the case
of an hs-boot file I think adding an import list is the right thing to
do regardless of the bug.

Previously it was awkwardly in TyCoFVs (and before that in TyCoRep).
Type seems like a sensible place for it to live.