GitLab

This reverts commit 2fc0c6cf.

This reverts commit e8dac6dc.

This updates `containers` to provide more specialised `Foldable`
instance methods, see also

  https://github.com/haskell/containers/issues/56

for more details.

Summary:
Get these lines fitting in 80 columns, and replace ptext (sLit ...) with text
Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Test Plan: validate

Reviewers: simonmar, austin

Subscribers: thomie, carter, ezyang, simonmar

Differential Revision: https://phabricator.haskell.org/D342

This commit removes a couple of

  {-# OPTIONS_GHC -fno-warn-unused-imports #-}

by cleaning up the imports, as well as ensuring that all modules in the
GHC.* hierarchy avoid importing the `Prelude` module to clean-up the
import graph a bit.

Summary:
This is a fixup of https://phabricator.haskell.org/D233

The only difference is in findTFiles (first commit), which
previously broke Windows runner; now I translated literally
instead attempting to improve it, and checked it works.

Test Plan:
I used validate under 2,3 on Linux and under 2 on msys2.
On Windows I've seen a large number of failures, but they don't
seem to be connected with the patch.

Reviewers: hvr, simonmar, thomie, austin

Reviewed By: austin

Subscribers: thomie, carter, ezyang, simonmar

Differential Revision: https://phabricator.haskell.org/D310

GHC Trac Issues: #9184

Since they're implied by the lack of `NoImplicitPrelude`

With this change `Control.Monad.foldM` becomes an alias for
`Data.Foldable.foldlM`.

Reviewed By: austin, ekmett

Differential Revision: https://phabricator.haskell.org/D251

This finally removes the `Data.OldTypeable` module (which
has been deprecated in 7.8), from `base`, compiler and testsuite.

The deprecated `Typeable{1..7}` aliases in `Data.Typeable` are not
removed yet in order to give existing code a bit more time to adapt.

Reviewed By: hvr, dreixel

Differential Revision: https://phabricator.haskell.org/D311

Summary:
This is an optimization to the CTZ primops introduced for #9340

Previously we called out to `hs_ctz64`, but we can actually generate
better hand-tuned code while avoiding the FFI ccall.

With this patch, the code

  {-# LANGUAGE MagicHash #-}
  module TestClz0 where
  import GHC.Prim
  ctz64 :: Word64# -> Word#
  ctz64 x = ctz64# x

results in the following assembler generated by NCG on i386:

  TestClz.ctz64_info:
      movl (%ebp),%eax
      movl 4(%ebp),%ecx
      movl %ecx,%edx
      orl %eax,%edx
      movl $64,%edx
      je _nAO

      bsf %ecx,%ecx
      addl $32,%ecx
      bsf %eax,%eax
      cmovne %eax,%ecx
      movl %ecx,%edx
  _nAO:
      movl %edx,%esi
      addl $8,%ebp
      jmp *(%ebp)

For comparision, here's what LLVM 3.4 currently generates:

  000000fc <TestClzz_ctzz64_info>:
    fc:   0f bc 45 04             bsf    0x4(%ebp),%eax
   100:   b9 20 00 00 00          mov    $0x20,%ecx
   105:   0f 45 c8                cmovne %eax,%ecx
   108:   83 c1 20                add    $0x20,%ecx
   10b:   8b 45 00                mov    0x0(%ebp),%eax
   10e:   8b 55 08                mov    0x8(%ebp),%edx
   111:   0f bc f0                bsf    %eax,%esi
   114:   85 c0                   test   %eax,%eax
   116:   0f 44 f1                cmove  %ecx,%esi
   119:   83 c5 08                add    $0x8,%ebp
   11c:   ff e2                   jmp    *%edx

Reviewed By: austin

Auditors: simonmar

Differential Revision: https://phabricator.haskell.org/D163

Summary:
If `pthread_setname_np` is not available, then a regular
./validate will fail due to warnings; the `name` parameter to
`createOSThread` becomes unused.
Signed-off-by: Austin Seipp <austin@well-typed.com>

Test Plan: iiam

Reviewers: simonmar, nomeata, jstolarek, hvr

Reviewed By: nomeata, jstolarek, hvr

Subscribers: nomeata, thomie, carter, ezyang, simonmar

Differential Revision: https://phabricator.haskell.org/D344

according to travis (and proably Harbormaster), although I just cannot
reproduce it here.

Akio found an avoidable cause of non-determinisim: The names of RULES
generated by Specialise had uniques in them:
    "SPEC $cshowsPrec_a2QX @ [GHC.Types.Char]" [ALWAYS] forall ...
By using showSDocForUser instead of showSDocDump when building the rule
name, this is avoided:
    "SPEC $cshowsPrec @ [Char]" [ALWAYS] forall ...
See #4012, comments 61ff.

Effective immediately, `deepseq` is maintained officially by the
core-library-comittee.  Moreover, pushing to libraries/deepseq requires
pushing to ssh://git@github.com/haskell/deepseq.git from now on.

[skip ci]

Reviewed By: nomeata, austin
Differential Revision: https://phabricator.haskell.org/D325

Fixes #9345. Use a modified banker's queue to achieve amortized optimal
performance for inits. The previous implementation was extremely slow.

Reviewed By: nomeata, ekmett, austin

Differential Revision: https://phabricator.haskell.org/D329

which I just put it in comments. Thanks Yuras for noticing, and sorry
for the noise.

to include at least what’s observed on ghc-speed and my laptop, and
hopefully also Harbormaster.

including T9675 itself. Maybe its memory behaviour is too spiky for
max_bytes_used and peak_megabytes_allocated to be used sensibly at all.

otherwise this can retain large lazy calculations. This fixed one space
leak pointed out in #9675.

although I obtained them originally from a clean validate tree on my
laptop, they did not match what I found on the build bot host.

so that whoever improves the situation can feel good about it.

See also https://github.com/haskell/time/issues/2

However, while the `time-1.5` package now loads successful in GHCi,
the linker warnings as reported in #9297 occur (which let the
testsuite fail for a dozen of testcases due to this additional
output)

This avoids the import-cycle caused by the import of `Foreign.C.Types`
by using `Int` instead of `CInt` for the Unicode classification
functions. This refactoring also allows to remove a couple of
`fromIntegral`s.

Reviewed By: rwbarton, ekmett

Differential Revision: https://phabricator.haskell.org/D328

Currently, the detection recognizes the following `uname -s` strings:

 - `CYGWIN_NT-6.3`
 - `MINGW32_NT-6.3`
 - `MINGW64_NT_6.3`

However, MSYS2 provides an additional target, in which case `uname -s`
returns a string such as `MSYS_NT-6.3`. In all these cases, the system
ought to be recognized as being a `windows` os by the testsuite runner.

See also #9604

On Windows, we may be using a native build of Python or a mingw/msys build. The
former exports `ctypes.windll`, the latter exports `cdll`. Previously the code
threw an exception when using the msys Python because it expected `windll` to
always be available on Windows.

Differential Revision: https://phabricator.haskell.org/D308

There don't seem to be any corresponding `{-# SOURCE #-}` for the removed
`.hs-boot`-files anymore (if there ever was any in the first place).

This also removes a commented out `{-# SOURCE #-}` import which turns up when
grepping the source for `{-# SOURCE #-}` occurences.

Caught by T6006 as a NULL dereference:

  Command: ./T6006

  Invalid read of size 8
     at 0x660ED9: hs_init_ghc (RtsStartup.c:168)
     by 0x660D90: hs_init (RtsStartup.c:112)
     by 0x40504D: main (in /home/slyfox/dev/git/ghc-validate/testsuite/tests/rts/T6006)
   Address 0x0 is not stack'd, malloc'd or (recently) free'd

The regression was introduced by commit
cb0a503aSigned-off-by: Sergei Trofimovich <slyfox@gentoo.org>

This helps identify threads in gdb particularly in processes with a
lot of threads.

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

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

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

Summary:
When linking fails because there was a problem with the supplied
object file, then we should not barf() or exit, we should emit a
suitable error message and return an error code to the caller.  We
should also free all memory that might have been allocated during
linking, and generally not do any damage.  This patch fixes most
common instances of this problem.

Test Plan: validate

Reviewers: rwbarton, austin, ezyang

Reviewed By: ezyang

Subscribers: simonmar, ezyang, carter, thomie

Differential Revision: https://phabricator.haskell.org/D294

Summary:
Rewrites takeWhile to a build/foldr form; fuses repeated
applications of takeWhile.

Reviewers: nomeata, austin

Reviewed By: nomeata

Subscribers: thomie, carter, ezyang, simonmar

Projects: #ghc

Differential Revision: https://phabricator.haskell.org/D322

GHC Trac Issues: #9132

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