ghc.git
5 years agoUpdate performance tests
Jan Stolarek [Fri, 18 Oct 2013 17:05:55 +0000 (19:05 +0200)] 
Update performance tests

Regressions in haddock for unknown reason

5 years agoUpdate performance test due to fix for #8456
Jan Stolarek [Fri, 18 Oct 2013 17:00:55 +0000 (19:00 +0200)] 
Update performance test due to fix for #8456

5 years agoT5313 now succeeds (Fixes: #7774)
Joachim Breitner [Fri, 18 Oct 2013 15:22:36 +0000 (17:22 +0200)] 
T5313 now succeeds (Fixes: #7774)

5 years agoTest Trac #8449
Simon Peyton Jones [Fri, 18 Oct 2013 11:30:19 +0000 (12:30 +0100)] 
Test Trac #8449

5 years agoAdded test for #7667.
Richard Eisenberg [Mon, 14 Oct 2013 21:39:48 +0000 (17:39 -0400)] 
Added test for #7667.

5 years agoTest #8412 (again)
Krzysztof Gogolewski [Sat, 12 Oct 2013 20:22:31 +0000 (22:22 +0200)] 
Test #8412 (again)

I forgot to git add a file.

5 years agoTest #8412
Krzysztof Gogolewski [Sat, 5 Oct 2013 15:23:59 +0000 (17:23 +0200)] 
Test #8412

5 years agoAdd a timeout for #8431
Krzysztof Gogolewski [Sat, 12 Oct 2013 03:05:32 +0000 (05:05 +0200)] 
Add a timeout for #8431

5 years agoTest #8431
Krzysztof Gogolewski [Sat, 12 Oct 2013 01:33:08 +0000 (03:33 +0200)] 
Test #8431

5 years agoTest #8430
Krzysztof Gogolewski [Fri, 11 Oct 2013 22:54:09 +0000 (00:54 +0200)] 
Test #8430

5 years agoUpdate test-suite for new `array-0.5.0.0`
Herbert Valerio Riedel [Fri, 11 Oct 2013 19:06:33 +0000 (21:06 +0200)] 
Update test-suite for new `array-0.5.0.0`

5 years agoaccept cgrun057(prof)
Simon Marlow [Fri, 11 Oct 2013 10:34:26 +0000 (11:34 +0100)] 
accept cgrun057(prof)

5 years agodon't need ghci_way_flags, the ghc package works all ways
Simon Marlow [Fri, 11 Oct 2013 09:36:52 +0000 (10:36 +0100)] 
don't need ghci_way_flags, the ghc package works all ways

5 years agoTest #8428
Krzysztof Gogolewski [Wed, 9 Oct 2013 20:10:50 +0000 (22:10 +0200)] 
Test #8428

5 years agoWibble after typofix
Krzysztof Gogolewski [Wed, 9 Oct 2013 18:04:11 +0000 (20:04 +0200)] 
Wibble after typofix

5 years agoWibble to annfail10 due to Edward's changes.
Austin Seipp [Wed, 9 Oct 2013 12:29:57 +0000 (07:29 -0500)] 
Wibble to annfail10 due to Edward's changes.

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoconcurrent/conc012: Pass RTS options to executable, not ghc
Ben Gamari [Fri, 20 Sep 2013 14:45:07 +0000 (10:45 -0400)] 
concurrent/conc012: Pass RTS options to executable, not ghc

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoError wibbles now that we print roles in :info
Simon Peyton Jones [Tue, 8 Oct 2013 17:08:31 +0000 (18:08 +0100)] 
Error wibbles now that we print roles in :info

(We might want to revisit this.)

5 years agoMore error-message wibbles
Simon Peyton Jones [Tue, 8 Oct 2013 12:12:12 +0000 (13:12 +0100)] 
More error-message wibbles

5 years agoChange deprecated _scc_ to {-# SCC #-} (#8170)
Krzysztof Gogolewski [Sat, 24 Aug 2013 14:56:14 +0000 (16:56 +0200)] 
Change deprecated _scc_ to {-# SCC #-} (#8170)

5 years agoAdd tests for new TH #4124, #4128, #4170, #4364, #6062
Krzysztof Gogolewski [Sat, 5 Oct 2013 17:08:09 +0000 (19:08 +0200)] 
Add tests for new TH #4124, #4128, #4170, #4364, #6062

5 years agoFlag to test suite: SKIP_PERF_TESTS
Joachim Breitner [Sat, 5 Oct 2013 15:04:31 +0000 (17:04 +0200)] 
Flag to test suite: SKIP_PERF_TESTS

More often than not the output of the performance tests is in the way,
rather than helping. This allows the use of `make SKIP_PERF_TESTS=YES`
to skip these tests. Fixes #8413

5 years agoMerge New Template Haskell branch.
Geoffrey Mainland [Fri, 4 Oct 2013 21:55:21 +0000 (17:55 -0400)] 
Merge New Template Haskell branch.

5 years agoPeculiarly, we now pass this test on ghci.
Geoffrey Mainland [Mon, 3 Jun 2013 14:00:58 +0000 (15:00 +0100)] 
Peculiarly, we now pass this test on ghci.

5 years agoFix test wibble.
Geoffrey Mainland [Wed, 22 May 2013 07:00:43 +0000 (08:00 +0100)] 
Fix test wibble.

5 years agoUpdate test to reflect fix for #1201.
Geoffrey Mainland [Wed, 22 May 2013 06:57:12 +0000 (07:57 +0100)] 
Update test to reflect fix for #1201.

Splices are now run in the renamer instead of the type checker, so they
only get run once.

5 years agoUpdate tests now that declaration splices are legal everywhere.
Geoffrey Mainland [Mon, 20 May 2013 10:38:54 +0000 (11:38 +0100)] 
Update tests now that declaration splices are legal everywhere.

5 years agoFix test wibbles for new Template Haskell.
Geoffrey Mainland [Thu, 16 May 2013 14:08:07 +0000 (15:08 +0100)] 
Fix test wibbles for new Template Haskell.

Because splices are now run in the renamer, we do not get the same error context
as we would when running in the type checker. In most cases we get less context,
and in some cases I have added additional context. Error messages should at
least tell the user that an error occurred in a splice; dropping context beyond
that point is not judged a great loss.

Note that we may now report only one error when multiple errors were reported
before because splices are now run in the renamer.

5 years agoAdjust tests for new Template Haskell.
Geoffrey Mainland [Thu, 16 May 2013 14:03:05 +0000 (15:03 +0100)] 
Adjust tests for new Template Haskell.

From the new Template Haskell proposal at
http://hackage.haskell.org/trac/ghc/blog/Template%20Haskell%20Proposal

  A declaration group is the chunk of declarations created by a top-level
  declaration splice, plus those following it, down to but not including the
  next top-level declaration splice. Then the type environment seen by reify
  includes all the declaration up to the end of the immediately preceding
  declaration block, but no more.

This change adds '$(return [])' where necessary to allow following declarations
to see (module-local) top-level definitions.

5 years agoError messsage wibblification
Simon Peyton Jones [Fri, 4 Oct 2013 17:55:13 +0000 (18:55 +0100)] 
Error messsage wibblification

5 years agoTest Trac #8383
Simon Peyton Jones [Fri, 4 Oct 2013 17:51:25 +0000 (18:51 +0100)] 
Test Trac #8383

5 years agoFix a failing test.
Iavor S. Diatchki [Fri, 4 Oct 2013 00:39:11 +0000 (17:39 -0700)] 
Fix a failing test.

GHC.TypeLits does not export `Sing` anymore, so we just define it
locally.

5 years agoFixup basic type-lits test.
Iavor S. Diatchki [Thu, 3 Oct 2013 23:28:16 +0000 (16:28 -0700)] 
Fixup basic type-lits test.

There is still one more test that needs fixing:

   indexed-types/should_fail  T7786 [stderr mismatch] (normal)

I need to understand what is going on there, as it appears to be
using the `Sing` constructors a bunch.

5 years agoNew tests for #8368.
Richard Eisenberg [Thu, 3 Oct 2013 14:22:27 +0000 (10:22 -0400)] 
New tests for #8368.

5 years agoTest Trac #8357
Simon Peyton Jones [Thu, 3 Oct 2013 14:40:13 +0000 (15:40 +0100)] 
Test Trac #8357

5 years agoTrac #T7881 now works for hpc way again
Simon Peyton Jones [Thu, 3 Oct 2013 11:16:15 +0000 (12:16 +0100)] 
Trac #T7881 now works for hpc way again

As a result of fixing Trac #8384

5 years agoTest Trac #8391
Simon Peyton Jones [Thu, 3 Oct 2013 11:10:38 +0000 (12:10 +0100)] 
Test Trac #8391

5 years agoTest Trac #8392
Simon Peyton Jones [Thu, 3 Oct 2013 08:25:25 +0000 (09:25 +0100)] 
Test Trac #8392

5 years agoError message wibbles following ambiguity check changes
Simon Peyton Jones [Wed, 2 Oct 2013 13:01:02 +0000 (14:01 +0100)] 
Error message wibbles following ambiguity check changes

5 years agoFurther testsuite wibbles following changes to kind suppression
Simon Peyton Jones [Wed, 2 Oct 2013 08:00:55 +0000 (09:00 +0100)] 
Further testsuite wibbles following changes to kind suppression

5 years agoAdd tests for prefetch primops (#8256)
Austin Seipp [Wed, 2 Oct 2013 02:20:24 +0000 (21:20 -0500)] 
Add tests for prefetch primops (#8256)

Authored-by: Carter Tazio Schonwald <carter.schonwald@gmail.com>
Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoAdd 32-bit output for T7919
Reid Barton [Wed, 2 Oct 2013 00:57:11 +0000 (20:57 -0400)] 
Add 32-bit output for T7919

Issue #8393

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoSkip integerGmpInternals in ghci (no unboxed tuples)
Reid Barton [Wed, 2 Oct 2013 00:56:20 +0000 (20:56 -0400)] 
Skip integerGmpInternals in ghci (no unboxed tuples)

Issue #8393

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoMark those tests which depend on libraries not built when cross-compiling
Reid Barton [Wed, 2 Oct 2013 00:56:02 +0000 (20:56 -0400)] 
Mark those tests which depend on libraries not built when cross-compiling

Issue #8393

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoBetter fundep error message
Simon Peyton Jones [Tue, 1 Oct 2013 17:27:21 +0000 (18:27 +0100)] 
Better fundep error message

5 years agoTest Trac #8359
Simon Peyton Jones [Tue, 1 Oct 2013 17:26:54 +0000 (18:26 +0100)] 
Test Trac #8359

5 years agoError message wibbles,
unknown [Tue, 1 Oct 2013 15:53:08 +0000 (16:53 +0100)] 
Error message wibbles,

following
  a) suppressing kind foralls and arguments
  b) better fundep error messages

5 years agoTest Trac #8329
Simon Peyton Jones [Tue, 1 Oct 2013 11:08:55 +0000 (12:08 +0100)] 
Test Trac #8329

5 years agoadd a sleep for systems with low-resolution file timestamps
Simon Marlow [Mon, 23 Sep 2013 11:17:13 +0000 (12:17 +0100)] 
add a sleep for systems with low-resolution file timestamps

5 years agonew test cases for MultiWayIf w/ layout
Simon Marlow [Mon, 30 Sep 2013 20:20:17 +0000 (21:20 +0100)] 
new test cases for MultiWayIf w/ layout

5 years agoAdd tests for new internal GMP primitives
Herbert Valerio Riedel [Tue, 1 Oct 2013 09:37:59 +0000 (11:37 +0200)] 
Add tests for new internal GMP primitives

See [ce8ec963/integer-gmp] and [97c101b7/integer-gmp] for more details

5 years agoFix T8129 (#8129)
Krzysztof Gogolewski [Tue, 1 Oct 2013 08:33:05 +0000 (10:33 +0200)] 
Fix T8129 (#8129)

5 years agoUpdate role annotation syntax in TcCoercible
Reid Barton [Sun, 29 Sep 2013 23:05:35 +0000 (19:05 -0400)] 
Update role annotation syntax in TcCoercible

5 years agoUpdate prog001 output (see #7633)
Reid Barton [Sun, 29 Sep 2013 22:56:53 +0000 (18:56 -0400)] 
Update prog001 output (see #7633)

5 years agoUpdate spec001 for new Typeable, and disable AMP warning
Reid Barton [Sun, 29 Sep 2013 22:42:35 +0000 (18:42 -0400)] 
Update spec001 for new Typeable, and disable AMP warning

I'd like to just "import Prelude hiding (..., join)", but the
patch to make that turn off the warning hasn't been merged yet.

5 years agoMark T7881 as expected failure for WAY=hpc (#8384)
Reid Barton [Sun, 29 Sep 2013 21:29:33 +0000 (17:29 -0400)] 
Mark T7881 as expected failure for WAY=hpc (#8384)

5 years agoUpdate compareAndSwap for new comparison primops
Reid Barton [Sun, 29 Sep 2013 20:36:54 +0000 (16:36 -0400)] 
Update compareAndSwap for new comparison primops

5 years agoDeal with failures for T367, T367_letnoescape under ghci
Reid Barton [Sun, 29 Sep 2013 20:21:01 +0000 (16:21 -0400)] 
Deal with failures for T367, T367_letnoescape under ghci

These tests had a very short timeout (0.3 s). With WAY=ghci,
the time ghci takes to start up and compile the test modules
is counted in this timeout, and that causes the tests to fail.

T367 really needs the very short timeout, so this commit disables
the ghci way for T367. T367_letnoescape can handle any timeout,
so I bumped up the timeout to 6 s to give ghci time to start up.

5 years agoFix fallout from making lazy unlifted bindings an error
Austin Seipp [Sun, 29 Sep 2013 22:22:22 +0000 (17:22 -0500)] 
Fix fallout from making lazy unlifted bindings an error

Issue #8022

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoTest Trac #8377
Takano Akio [Sat, 28 Sep 2013 11:05:33 +0000 (20:05 +0900)] 
Test Trac #8377

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoFix T8333 (#8333)
Krzysztof Gogolewski [Sat, 28 Sep 2013 10:58:40 +0000 (12:58 +0200)] 
Fix T8333 (#8333)

5 years agoFix T8132
Austin Seipp [Fri, 27 Sep 2013 05:00:17 +0000 (00:00 -0500)] 
Fix T8132

A tiny bit of fallout from the TypeRep changes with a simple fix: we
don't export typeRep# from Data.Typeable anymore, only
Data.Typeable.Internal

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoAdd some tests for type-level subtraction.
Iavor S. Diatchki [Tue, 24 Sep 2013 13:14:53 +0000 (06:14 -0700)] 
Add some tests for type-level subtraction.

5 years agoDriver: Fix % operator typo
Joachim Breitner [Mon, 23 Sep 2013 20:58:15 +0000 (22:58 +0200)] 
Driver: Fix % operator typo

5 years agoWibble after #8218
Krzysztof Gogolewski [Mon, 23 Sep 2013 15:57:31 +0000 (17:57 +0200)] 
Wibble after #8218

The new hidden line in output of tcfail181 is

instance Monad m => Monad (Control.Applicative.WrappedMonad m).

It might seem weird that Control.Applicative is suggested when
not imported, but it's almost certainly a side-effect of AMP and
it's harmless.

5 years agolinker_unload is now passing (#8192)
Simon Marlow [Sun, 22 Sep 2013 15:49:07 +0000 (16:49 +0100)] 
linker_unload is now passing (#8192)

5 years agoT7574 is now passing (#7574)
Simon Marlow [Sun, 22 Sep 2013 15:48:50 +0000 (16:48 +0100)] 
T7574 is now passing (#7574)

5 years agoTypos
Krzysztof Gogolewski [Mon, 23 Sep 2013 09:32:06 +0000 (11:32 +0200)] 
Typos

5 years agoTest Trac #8333
Simon Peyton Jones [Fri, 20 Sep 2013 14:35:21 +0000 (15:35 +0100)] 
Test Trac #8333

5 years agotest Trac #7800
Simon Peyton Jones [Fri, 20 Sep 2013 14:30:03 +0000 (15:30 +0100)] 
test Trac #7800

5 years agoTest Trac #8129
Simon Peyton Jones [Thu, 19 Sep 2013 12:21:21 +0000 (13:21 +0100)] 
Test Trac #8129

5 years agoTest Trac #8280
Simon Peyton Jones [Wed, 18 Sep 2013 15:25:51 +0000 (16:25 +0100)] 
Test Trac #8280

5 years agoTest driver: Implement different verbosities
Joachim Breitner [Fri, 20 Sep 2013 13:15:51 +0000 (15:15 +0200)] 
Test driver: Implement different verbosities

Select verbosity with "make VERBOSE=n". Options so far:
 n=0: No per-test output
 n=1: Only failing test results
 n=2: As above, plus progress information (names of all tests)
 n=3: As aobve, plus commands called.
Default currently is n=3, although n=2 might be a nicer default.

5 years agoTest driver: Print total time spent at the end
Joachim Breitner [Fri, 20 Sep 2013 11:16:53 +0000 (13:16 +0200)] 
Test driver: Print total time spent at the end

5 years agoFix performance failures on x86_64
Jan Stolarek [Wed, 18 Sep 2013 16:58:38 +0000 (17:58 +0100)] 
Fix performance failures on x86_64

5 years agoFix T4007
Jan Stolarek [Wed, 18 Sep 2013 16:18:33 +0000 (17:18 +0100)] 
Fix T4007

Accidentally broken by my last commit

5 years agoFollow changes in comparison primops (see #6135)
Jan Stolarek [Mon, 16 Sep 2013 14:15:59 +0000 (15:15 +0100)] 
Follow changes in comparison primops (see #6135)

5 years agoTrailing whitespaces
Jan Stolarek [Mon, 16 Sep 2013 15:05:37 +0000 (16:05 +0100)] 
Trailing whitespaces

5 years agoFollow changes in [7ab62490/base] (re #7633)
Herbert Valerio Riedel [Wed, 18 Sep 2013 11:27:14 +0000 (13:27 +0200)] 
Follow changes in [7ab62490/base] (re #7633)

5 years agoFollow changes in [12a8244/ghc-prim] (#7633)
Herbert Valerio Riedel [Wed, 18 Sep 2013 08:40:12 +0000 (10:40 +0200)] 
Follow changes in [12a8244/ghc-prim] (#7633)

5 years agoTests for MINIMAL pragma (#7633)
Twan van Laarhoven [Thu, 12 Sep 2013 17:13:48 +0000 (19:13 +0200)] 
Tests for MINIMAL pragma (#7633)

5 years agoUpdate to new missing-method-decl warning style
Herbert Valerio Riedel [Tue, 17 Sep 2013 21:57:42 +0000 (23:57 +0200)] 
Update to new missing-method-decl warning style

As a side-effect, the new `{-# MINIMAL #-}` facility (see #7633) changes
the existing separate warnings for each missing method declaration:

{{{
No explicit method or default declaration for ‛+’
In the instance declaration for ‛Num T’
}}}

to a single warning per class of the following form:

{{{
No explicit implementation for
  ‛+’, ‛*’, ‛abs’, ‛signum’, and ‛fromInteger’
In the instance declaration for ‛Num T’
}}}

This commit adapts the reference output files in the testsuite to follow
this change in GHC's warnings messages.

Signed-off-by: Herbert Valerio Riedel <hvr@gnu.org>
5 years agoUpdate to tests due to change in syntax for role annotations.
Richard Eisenberg [Wed, 28 Aug 2013 16:05:03 +0000 (12:05 -0400)] 
Update to tests due to change in syntax for role annotations.

5 years agoAdd missing Windows stdout/stderr output for #5435.
Edward Z. Yang [Tue, 17 Sep 2013 22:51:22 +0000 (15:51 -0700)] 
Add missing Windows stdout/stderr output for #5435.

Signed-off-by: Edward Z. Yang <ezyang@mit.edu>
5 years agoRecompilation test case for #8247
Joachim Breitner [Tue, 17 Sep 2013 20:37:51 +0000 (22:37 +0200)] 
Recompilation test case for #8247

But seems to be fixed in head already.

5 years agoAdjust testsuite to new printf implementation
Joachim Breitner [Tue, 17 Sep 2013 10:21:47 +0000 (12:21 +0200)] 
Adjust testsuite to new printf implementation

Expand 'bytes allocated' allowed in haddock.base, and update error
messages.

5 years agoIn the test that tests -optP -C, also pass -optP -ffreestanding
Joachim Breitner [Tue, 17 Sep 2013 13:47:11 +0000 (15:47 +0200)] 
In the test that tests -optP -C, also pass -optP -ffreestanding

This closes #3389, and makes the testsuite go through again.

5 years agoFinish up asm test #5435 for Mac OS X
Edward Z. Yang [Tue, 17 Sep 2013 07:02:45 +0000 (00:02 -0700)] 
Finish up asm test #5435 for Mac OS X

Signed-off-by: Edward Z. Yang <ezyang@mit.edu>
5 years agoImplement assembly test for #5435 in Windows.
Edward Z. Yang [Tue, 17 Sep 2013 00:59:39 +0000 (17:59 -0700)] 
Implement assembly test for #5435 in Windows.

Fixes HostOS bug, where the define was not being set.

Signed-off-by: Edward Z. Yang <ezyang@mit.edu>
5 years agoNew and improved tests for #5435. Linux only at the moment.
Edward Z. Yang [Tue, 17 Sep 2013 00:05:10 +0000 (17:05 -0700)] 
New and improved tests for #5435. Linux only at the moment.

Added a new test Makefile variable $(dllext), which is
instantiated to .dll or .so or .dylib depending on your platform.

Signed-off-by: Edward Z. Yang <ezyang@mit.edu>
5 years agoTest for #8306
Austin Seipp [Mon, 16 Sep 2013 15:45:00 +0000 (10:45 -0500)] 
Test for #8306

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoT5435 test improvements, see #5435 for details.
Edward Z. Yang [Sun, 15 Sep 2013 21:07:59 +0000 (14:07 -0700)] 
T5435 test improvements, see #5435 for details.

Signed-off-by: Edward Z. Yang <ezyang@mit.edu>
5 years agoAdd test case for #8184
Patrick Palka [Sun, 15 Sep 2013 20:16:06 +0000 (16:16 -0400)] 
Add test case for #8184

5 years agoFix usage of 'expect_fail' in T2507
Patrick Palka [Sun, 15 Sep 2013 20:21:42 +0000 (16:21 -0400)] 
Fix usage of 'expect_fail' in T2507

5 years agoDelete trailing whitespace in cgrun070
Austin Seipp [Sun, 15 Sep 2013 18:54:33 +0000 (13:54 -0500)] 
Delete trailing whitespace in cgrun070

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoFix up cgrun072 a bit (#8250)
Austin Seipp [Sun, 15 Sep 2013 18:01:38 +0000 (13:01 -0500)] 
Fix up cgrun072 a bit (#8250)

This includes:

 * Adding a test for bswap16 with a low byte >= 128
 * Also test the byteSwapN functions from GHC.Word, tested both INLINE
   and not INLINE, so we test both independent parts: the compilation of
   base, and the backend compiling the code *using* base.
 * Fix the usage of byteSwapN# primitives in the test, by masking off
 * the higher bits when storing the results in Word16/Word32.

Thanks to Reid Barton for the investigation.

Authored-by: Reid Barton <rwbarton@gmail.com>
Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoAdd tests for the new ByteArray# <-> Addr# copy primops
Duncan Coutts [Sat, 14 Sep 2013 09:32:30 +0000 (10:32 +0100)] 
Add tests for the new ByteArray# <-> Addr# copy primops

Essentially the same tests as for the existing ByteArray# ones.

Signed-off-by: Austin Seipp <austin@well-typed.com>
5 years agoMake these two tests thread safe.
Edward Z. Yang [Sat, 14 Sep 2013 20:33:52 +0000 (13:33 -0700)] 
Make these two tests thread safe.

Signed-off-by: Edward Z. Yang <ezyang@mit.edu>
5 years agoAdjust test suite to new Language Pragma warnigns
Joachim Breitner [Fri, 13 Sep 2013 11:51:33 +0000 (13:51 +0200)] 
Adjust test suite to new Language Pragma warnigns

(this is related to #3647)

5 years agoSuppress some known broken Windows tests.
Edward Z. Yang [Sat, 14 Sep 2013 08:56:02 +0000 (01:56 -0700)] 
Suppress some known broken Windows tests.

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