Previously programs only depended upon the direct dependencies; while I
would have thought that this would be sufficient, somehow we were
getting to the link step of building `ghc-pkg` before `ghc-boot-th` was
built (despite the fact that `ghc-boot` has a direct dependency on
`ghc-boot-th`).

See #12078.

(cherry picked from commit 5d80d141)

There was a missing `#`. Hadrian couldn't come soon enough.

(cherry picked from commit a54d87a9)

This may have been the cause of various build failures on FreeBSD and
Solaris.

(cherry picked from commit 770d7085)

Remove html_theme declaration.

Simon Marlow authored 8 years ago and Ben Gamari committed 8 years ago

After a parallel GC, it is possible to have a long list of blocks in
ws->part_list, if we did a lot of work stealing but didn't fill up the
blocks we stole.  These blocks persist until the next load-balanced GC,
which might be a long time, and during every GC we were traversing this
list to find its size.  The fix is to maintain the size all the time, so
we don't have to compute it.

(cherry picked from commit 5c4cd0e4)

This will be the 2.17.2 haddock release.

This isn't supported by GHC 7.8

Ben Gamari authored 8 years ago and Ben Gamari committed 8 years ago

This creates a new package, `ghc-boot-th`, to contain the `Extension`
type, which now lives in `GHC.LanguageExtension.Type`. This ensures that
the transitive dependency set of the `template-haskell` package remains
minimal.

The `GHC.LanguageExtensions.Type` module is also re-exported by
`ghc-boot`, which provides an orphan `binary` instance as well.

Test Plan: Validate

Reviewers: goldfire, thomie, hvr, austin

Reviewed By: thomie

Subscribers: RyanGlScott, thomie, erikd, ezyang

Differential Revision: https://phabricator.haskell.org/D2224

(cherry picked from commit eed820b6)

(cherry picked from commit d78faa13)

Peter Trommler authored 8 years ago and Ben Gamari committed 8 years ago

The SMP primitives were missing appropriate memory barriers
(sync, isync instructions) on all PowerPCs.

Use the built-ins _sync_* provided by gcc and clang. This
reduces code size significantly.

Remove broken mark for concprog001 on powerpc64. The referenced
ticket number (11259) was wrong.

Test Plan: validate on powerpc and ARM

Reviewers: erikd, austin, simonmar, bgamari, hvr

Reviewed By: bgamari, hvr

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D2225

GHC Trac Issues: #12070

(cherry picked from commit 563a4857)

Peter Trommler authored 8 years ago and Ben Gamari committed 8 years ago

Printing STU was mixed up. The tab character must appear
after the 'x'.

Test Plan: validate on powerpc

Reviewers: bgamari, austin, erikd

Reviewed By: austin, erikd

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D2214

GHC Trac Issues: #12054

(cherry picked from commit 2dbdc79b)

This resolves #12058.

Thanks to Páli for tidentifying this issue so clearly.

Fixes `make clean`.

Ensures that haddock documentation is built as part of the `all` target.

takano-akio authored 8 years ago and Ben Gamari committed 8 years ago

Previously, we ignored promotion failures when evacuating fields of
a WEAK object. When a failure happens, this resulted in an WEAK object
pointing to another object in a younger generation, causing crashes.

I used the test case from #11746 to check that the fix is working.
However I haven't managed to produce a test case that quickly reproduces
the issue.

Test Plan: ./validate

Reviewers: austin, bgamari, simonmar

Reviewed By: simonmar

Subscribers: thomie

Differential Revision: https://phabricator.haskell.org/D2189

GHC Trac Issues: #11108

(cherry picked from commit 9363f04d)

Thanks Mikhail!

This suggestion only applies to operators.

Matthew Pickering authored 8 years ago and Ben Gamari committed 8 years ago

In the long discussion on #11432, it was decided that when a type
constructor is parsed as a variable ((--.->) is one example) then in
order to export the type constructor then the user should be required to
use the ExplicitNamespaces keyword.

This was implemented in quite an indirect manner in the renamer. It
is much more direct to enforce this in the parser at the expense of
slighty worse error messages. Further to this, the check in the
renamer was actually slightly wrong. If the variable was in scope
then no error was raised, this was causing panics, see #12026 for an
example.

Reviewers: austin, bgamari

Subscribers: davean, skvadrik, thomie

Differential Revision: https://phabricator.haskell.org/D2181

GHC Trac Issues: #12026

Mention -fshow-warning-groups, etc.

Herbert Valerio Riedel authored 8 years ago and Ben Gamari committed 8 years ago

(cherry picked from commit 633b099c)

This isn't quite a valid release yet but at least it's closer.

(cherry picked from commit 4466ae64)

(cherry picked from commit 0051ac17)