ghc.git
6 years agoImplement poly-kinded Typeable
Jose Pedro Magalhaes [Thu, 7 Feb 2013 13:59:51 +0000 (13:59 +0000)] 
Implement poly-kinded Typeable

This patch makes the Data.Typeable.Typeable class work with arguments of any
kind. In particular, this removes the Typeable1..7 class hierarchy, greatly
simplyfing the whole Typeable story. Also added is the AutoDeriveTypeable
language extension, which will automatically derive Typeable for all types and
classes declared in that module. Since there is now no good reason to give
handwritten instances of the Typeable class, those are ignored (for backwards
compatibility), and a warning is emitted.

The old, kind-* Typeable class is now called OldTypeable, and lives in the
Data.OldTypeable module. It is deprecated, and should be removed in some future
version of GHC.

6 years agoUpdate test to match new I/O manager behavior
Johan Tibell [Sat, 9 Feb 2013 00:59:16 +0000 (16:59 -0800)] 
Update test to match new I/O manager behavior

The new I/O manager has a separate thread for tracking timeouts.

6 years agoAdded testcase for Trac #7681, a Template Haskell missing feature.
Richard Eisenberg [Tue, 12 Feb 2013 04:06:41 +0000 (23:06 -0500)] 
Added testcase for Trac #7681, a Template Haskell missing feature.

6 years agoFix result001
Ian Lynagh [Mon, 11 Feb 2013 19:06:33 +0000 (19:06 +0000)] 
Fix result001

6 years agoprint020 now passes
Ian Lynagh [Mon, 11 Feb 2013 18:54:37 +0000 (18:54 +0000)] 
print020 now passes

6 years agoUpdate print020 expected output
Ian Lynagh [Mon, 11 Feb 2013 18:53:04 +0000 (18:53 +0000)] 
Update print020 expected output

6 years agoHappyTest needs MagicHash
Ian Lynagh [Mon, 11 Feb 2013 18:51:36 +0000 (18:51 +0000)] 
HappyTest needs MagicHash

6 years agoAdd expected warnings for print020
Ian Lynagh [Mon, 11 Feb 2013 18:51:27 +0000 (18:51 +0000)] 
Add expected warnings for print020

6 years agoUpdate the reason for T5267 being broken
Ian Lynagh [Mon, 11 Feb 2013 18:35:02 +0000 (18:35 +0000)] 
Update the reason for T5267 being broken

6 years agoRemove ds061
Ian Lynagh [Mon, 11 Feb 2013 18:32:34 +0000 (18:32 +0000)] 
Remove ds061

It was a failing test for a defunct feature (NPlusKPatterns) (#851)

6 years agoAdd expected output for T6117, which is now working
Ian Lynagh [Mon, 11 Feb 2013 18:30:24 +0000 (18:30 +0000)] 
Add expected output for T6117, which is now working

6 years agoRemove T5763
Ian Lynagh [Mon, 11 Feb 2013 18:22:02 +0000 (18:22 +0000)] 
Remove T5763

It was added in fbb1f167657bcdb4d9a67e9b97734faeb82c8a25 without
an accompanying source file. Also, it was marked as being broken by
5673, so I'm not even sure which ticket it was supposed to be a
test for.

6 years agoTweak the brokens list to include the directory the test is in
Ian Lynagh [Mon, 11 Feb 2013 17:36:40 +0000 (17:36 +0000)] 
Tweak the brokens list to include the directory the test is in

6 years agoTicket #1845 is closed, so ghci024 presumably works on PPC/OSX now
Ian Lynagh [Mon, 11 Feb 2013 17:36:06 +0000 (17:36 +0000)] 
Ticket #1845 is closed, so ghci024 presumably works on PPC/OSX now

6 years agoMerge branch 'master' of darcs.haskell.org:/srv/darcs//testsuite
Ian Lynagh [Mon, 11 Feb 2013 15:53:13 +0000 (15:53 +0000)] 
Merge branch 'master' of darcs.haskell.org:/srv/darcs//testsuite

6 years agoMore conversions
Ian Lynagh [Mon, 11 Feb 2013 15:47:02 +0000 (15:47 +0000)] 
More conversions

6 years agoMore helper conversions
Ian Lynagh [Mon, 11 Feb 2013 15:39:12 +0000 (15:39 +0000)] 
More helper conversions

6 years agoConvert more helpers
Ian Lynagh [Mon, 11 Feb 2013 15:11:52 +0000 (15:11 +0000)] 
Convert more helpers

6 years agoConvert more helper functions
Ian Lynagh [Mon, 11 Feb 2013 15:07:12 +0000 (15:07 +0000)] 
Convert more helper functions

6 years agoConvert more helper functions to the new scheme
Ian Lynagh [Mon, 11 Feb 2013 14:18:38 +0000 (14:18 +0000)] 
Convert more helper functions to the new scheme

6 years agoConvert more helpers to the new scheme
Ian Lynagh [Mon, 11 Feb 2013 14:08:58 +0000 (14:08 +0000)] 
Convert more helpers to the new scheme

6 years agoChange '{if,unless}_arch' to 'arch'
Ian Lynagh [Mon, 11 Feb 2013 14:03:23 +0000 (14:03 +0000)] 
Change '{if,unless}_arch' to 'arch'

6 years agoPrint a warning if we get framework failures when listing brokens
Ian Lynagh [Mon, 11 Feb 2013 13:59:55 +0000 (13:59 +0000)] 
Print a warning if we get framework failures when listing brokens

6 years agoHandle duplicates in the brokens list
Ian Lynagh [Mon, 11 Feb 2013 13:55:40 +0000 (13:55 +0000)] 
Handle duplicates in the brokens list

6 years agoexpect_broken_for should also be added to the list of brokens
Ian Lynagh [Mon, 11 Feb 2013 13:52:49 +0000 (13:52 +0000)] 
expect_broken_for should also be added to the list of brokens

6 years agoReplace 'if_os' and 'unless_os' with 'opsys'
Ian Lynagh [Mon, 11 Feb 2013 13:48:59 +0000 (13:48 +0000)] 
Replace 'if_os' and 'unless_os' with 'opsys'

The name 'os' clashes with a python library

6 years agoAdd 'make list_brokens'
Ian Lynagh [Mon, 11 Feb 2013 13:27:58 +0000 (13:27 +0000)] 
Add 'make list_brokens'

Gives a list of tickets that the testsuite thinks are broken, and
what bug it thinks is the reason. This can then be pasted into trac
and 'previewed', which will show any closed tickets with strikeout.

6 years agoMerge branch 'master' of http://darcs.haskell.org/testsuite
Simon Peyton Jones [Mon, 11 Feb 2013 08:45:25 +0000 (08:45 +0000)] 
Merge branch 'master' of darcs.haskell.org/testsuite

6 years agoT5113 is fixed
Simon Peyton Jones [Mon, 11 Feb 2013 08:45:01 +0000 (08:45 +0000)] 
T5113 is fixed

6 years agoImprove bound for T3064
Simon Peyton Jones [Mon, 11 Feb 2013 08:43:40 +0000 (08:43 +0000)] 
Improve bound for T3064

6 years agoTweak another 32bit perf bound
Ian Lynagh [Sun, 10 Feb 2013 17:54:30 +0000 (17:54 +0000)] 
Tweak another 32bit perf bound

6 years agoTweak another 32bit perf bound
Ian Lynagh [Sun, 10 Feb 2013 17:53:02 +0000 (17:53 +0000)] 
Tweak another 32bit perf bound

6 years agoMore 32bit perf bound fixes
Ian Lynagh [Sun, 10 Feb 2013 17:45:59 +0000 (17:45 +0000)] 
More 32bit perf bound fixes

6 years agoMore 32bit perf test bound updates
Ian Lynagh [Sun, 10 Feb 2013 17:33:53 +0000 (17:33 +0000)] 
More 32bit perf test bound updates

6 years agoMore 32-bit perf updates
Ian Lynagh [Sun, 10 Feb 2013 17:05:21 +0000 (17:05 +0000)] 
More 32-bit perf updates

6 years agoRearrange the summary output
Ian Lynagh [Sun, 10 Feb 2013 16:48:24 +0000 (16:48 +0000)] 
Rearrange the summary output

In particular, the 3 values you most need to care about (framework
failures, unexpected passes, unexpected failures) are now together
in a block.

6 years agoUpdate 32bit perf results for the haddock tests
Ian Lynagh [Sun, 10 Feb 2013 16:45:24 +0000 (16:45 +0000)] 
Update 32bit perf results for the haddock tests

6 years agoMerge branch 'master' of http://darcs.haskell.org/testsuite
Simon Peyton Jones [Fri, 8 Feb 2013 17:30:46 +0000 (17:30 +0000)] 
Merge branch 'master' of darcs.haskell.org/testsuite

6 years agoRemove unused helper functions
Ian Lynagh [Fri, 8 Feb 2013 13:36:22 +0000 (13:36 +0000)] 
Remove unused helper functions

6 years agoConvert remaining *_range_field's to *_num_field
Ian Lynagh [Fri, 8 Feb 2013 13:35:34 +0000 (13:35 +0000)] 
Convert remaining *_range_field's to *_num_field

6 years agoMerge branch 'master' of http://darcs.haskell.org/testsuite
Simon Peyton Jones [Fri, 8 Feb 2013 12:34:14 +0000 (12:34 +0000)] 
Merge branch 'master' of darcs.haskell.org/testsuite

6 years agoSwitch some more tests to use the new helpers
Ian Lynagh [Fri, 8 Feb 2013 00:54:41 +0000 (00:54 +0000)] 
Switch some more tests to use the new helpers

6 years agoReplace 'if_platform' and 'unless_platform' with 'platform'
Ian Lynagh [Fri, 8 Feb 2013 00:38:42 +0000 (00:38 +0000)] 
Replace 'if_platform' and 'unless_platform' with 'platform'

6 years agoConvert to stats_num_field in perf/should_run
Ian Lynagh [Thu, 7 Feb 2013 23:58:38 +0000 (23:58 +0000)] 
Convert to stats_num_field in perf/should_run

6 years agoAllow a simpler form of stats_num_field where all platforms use the same value
Ian Lynagh [Thu, 7 Feb 2013 23:36:55 +0000 (23:36 +0000)] 
Allow a simpler form of stats_num_field where all platforms use the same value

6 years agoAdd a different sort of stats_num_field helper function
Ian Lynagh [Thu, 7 Feb 2013 22:47:30 +0000 (22:47 +0000)] 
Add a different sort of stats_num_field helper function

Uses look like
    stats_num_field('bytes allocated',
                    [(wordsize(32), 45648, 5),
                     (wordsize(64), 49400, 5)])
where the first matching triple will be used. e.g. we could override
the Win32 expected values with:
                    [(platform('i386-unknown-mingw32'), 41000, 5),
                     (wordsize(32),                     45648, 5),
                     (wordsize(64),                     49400, 5)])
with other 32-bit platforms falling through to the wordsize(32) case.

This makes it easier to give different values for different platforms,
while being sure that all platforms are covered.

6 years agoRemove unused *_num_fields bindings
Ian Lynagh [Thu, 7 Feb 2013 21:39:54 +0000 (21:39 +0000)] 
Remove unused *_num_fields bindings

6 years agoPass the test name to the test options
Ian Lynagh [Thu, 7 Feb 2013 21:37:15 +0000 (21:37 +0000)] 
Pass the test name to the test options

This allows them to give framework failures.

I also had to change how setTestOpts works. Now, rather than applying
the options to the directory's "default options", it just stores the
options to be applied for each test (i.e. once we know the test name).

6 years agoChange how the testsuite driver handles copying the test options
Ian Lynagh [Thu, 7 Feb 2013 19:00:51 +0000 (19:00 +0000)] 
Change how the testsuite driver handles copying the test options

We now deepcopy the options for each test, which means that the test
helpers don't need to worry about sharing.

6 years agoRename some arguments
Ian Lynagh [Thu, 7 Feb 2013 18:19:26 +0000 (18:19 +0000)] 
Rename some arguments

6 years agoDefine 'when' and 'unless' helpers
Ian Lynagh [Thu, 7 Feb 2013 18:12:41 +0000 (18:12 +0000)] 
Define 'when' and 'unless' helpers

This will reduce the number of helper functions that we need

6 years agoSimplify driver code slightly
Ian Lynagh [Thu, 7 Feb 2013 17:01:46 +0000 (17:01 +0000)] 
Simplify driver code slightly

6 years agoAdd a TEST="..." line to testsuite output
Ian Lynagh [Thu, 7 Feb 2013 16:59:28 +0000 (16:59 +0000)] 
Add a TEST="..." line to testsuite output

This gives hte list of tests with unexpected results, so that you can
easily run them again.

6 years agoadd test for #7636
Simon Marlow [Thu, 7 Feb 2013 09:52:20 +0000 (09:52 +0000)] 
add test for #7636

6 years agoFix perf bounds on amd64/Linux following the change to *_range_field
Ian Lynagh [Thu, 7 Feb 2013 02:02:53 +0000 (02:02 +0000)] 
Fix perf bounds on amd64/Linux following the change to *_range_field

6 years agoMake failing perf test output easier to read
Ian Lynagh [Wed, 6 Feb 2013 17:30:23 +0000 (17:30 +0000)] 
Make failing perf test output easier to read

6 years agoEliminate *_num_field
Ian Lynagh [Wed, 6 Feb 2013 15:59:09 +0000 (15:59 +0000)] 
Eliminate *_num_field

We now use *_range_field everywhere instead

6 years agoFix framework failure
Ian Lynagh [Wed, 6 Feb 2013 14:33:37 +0000 (14:33 +0000)] 
Fix framework failure

6 years agodph/classes: fails for the moment, but earlier success was spurious
Manuel M T Chakravarty [Tue, 5 Feb 2013 07:10:38 +0000 (18:10 +1100)] 
dph/classes: fails for the moment, but earlier success was spurious

6 years agodph: tests with dph-lifted-copy need to use '-fno-vectorisation-avoidance if scalar...
Manuel M T Chakravarty [Tue, 5 Feb 2013 06:55:47 +0000 (17:55 +1100)] 
dph: tests with dph-lifted-copy need to use '-fno-vectorisation-avoidance if scalar closures get too big

6 years agodph/words: don't inline the vectorised exported functions
Manuel M T Chakravarty [Mon, 4 Feb 2013 23:32:10 +0000 (10:32 +1100)] 
dph/words: don't inline the vectorised exported functions

6 years agoFix safePkg01 on Windows
Ian Lynagh [Tue, 5 Feb 2013 21:56:27 +0000 (21:56 +0000)] 
Fix safePkg01 on Windows

6 years agoFix T7037 on Win64
Ian Lynagh [Tue, 5 Feb 2013 21:34:59 +0000 (21:34 +0000)] 
Fix T7037 on Win64

6 years agoUse normalise_errmsg_fun rather than normalise_fun where appropriate
Ian Lynagh [Tue, 5 Feb 2013 20:17:58 +0000 (20:17 +0000)] 
Use normalise_errmsg_fun rather than normalise_fun where appropriate

6 years agodynamicToo001 is broken on Windows (#7665)
Ian Lynagh [Tue, 5 Feb 2013 20:08:05 +0000 (20:08 +0000)] 
dynamicToo001 is broken on Windows (#7665)

6 years agoFix cabal01 on Windows
Ian Lynagh [Tue, 5 Feb 2013 19:53:20 +0000 (19:53 +0000)] 
Fix cabal01 on Windows

We were getting a
    Creating library file: dist\build\libHStest-1.0-ghc7.7.20130205.dll.a
message on stderr. We can't turn it off, so now the test just normalises
it away.

6 years agoSkip ghci/linking tests if we don't have ghci.
Austin Seipp [Mon, 4 Feb 2013 03:22:51 +0000 (21:22 -0600)] 
Skip ghci/linking tests if we don't have ghci.

Otherwise, you get annoying failures if you run 'make fast stage=1'.

Signed-off-by: Austin Seipp <mad.one@gmail.com>
6 years agoTidy up some tests
Ian Lynagh [Sun, 3 Feb 2013 19:25:54 +0000 (19:25 +0000)] 
Tidy up some tests

We now use pre_cmd rather than cmd_prefix wherever possible.
Also, pass "-s --no-print-directory" whenever we use pre_cmd.

6 years agoUse a proper executable (rather than a shell script) in T6106
Ian Lynagh [Sun, 3 Feb 2013 18:41:55 +0000 (18:41 +0000)] 
Use a proper executable (rather than a shell script) in T6106

Fixes the test on Windows

6 years agoFix shared001
Ian Lynagh [Sun, 3 Feb 2013 16:21:43 +0000 (16:21 +0000)] 
Fix shared001

6 years agoHandle ^C better when threads are being used too
Ian Lynagh [Sat, 2 Feb 2013 19:53:46 +0000 (19:53 +0000)] 
Handle ^C better when threads are being used too

6 years agoWhen the testsuite is ^Ced, print the summary anyway
Ian Lynagh [Sat, 2 Feb 2013 19:04:18 +0000 (19:04 +0000)] 
When the testsuite is ^Ced, print the summary anyway

6 years agoAdd a test for #1849
Ian Lynagh [Sat, 2 Feb 2013 16:34:53 +0000 (16:34 +0000)] 
Add a test for #1849

6 years agospelling
Gabor Greif [Fri, 1 Feb 2013 13:52:06 +0000 (14:52 +0100)] 
spelling

6 years agoT5113 is failing again
Ian Lynagh [Fri, 1 Feb 2013 20:41:37 +0000 (20:41 +0000)] 
T5113 is failing again

6 years agoAdd a test for #2222
Ian Lynagh [Fri, 1 Feb 2013 18:15:40 +0000 (18:15 +0000)] 
Add a test for #2222

6 years agoAccept output for readFail007
Ian Lynagh [Fri, 1 Feb 2013 15:34:09 +0000 (15:34 +0000)] 
Accept output for readFail007

6 years agoAdd a test for #984
Ian Lynagh [Fri, 1 Feb 2013 15:26:06 +0000 (15:26 +0000)] 
Add a test for #984

6 years agoMerge branch 'master' of http://darcs.haskell.org/testsuite
Simon Peyton Jones [Wed, 30 Jan 2013 14:31:19 +0000 (14:31 +0000)] 
Merge branch 'master' of darcs.haskell.org/testsuite

6 years agoError messsage wibbles
Simon Peyton Jones [Wed, 30 Jan 2013 10:03:39 +0000 (10:03 +0000)] 
Error messsage wibbles

6 years agoTest Trac #7601
Simon Peyton Jones [Wed, 30 Jan 2013 08:32:40 +0000 (08:32 +0000)] 
Test Trac #7601

6 years agoMerge branch 'master' of http://darcs.haskell.org/testsuite
Simon Peyton Jones [Wed, 30 Jan 2013 08:30:15 +0000 (08:30 +0000)] 
Merge branch 'master' of darcs.haskell.org/testsuite

6 years agoTest Trac #7282
Simon Peyton Jones [Wed, 30 Jan 2013 08:29:15 +0000 (08:29 +0000)] 
Test Trac #7282

6 years agoImprove test
Simon Peyton Jones [Wed, 30 Jan 2013 08:26:33 +0000 (08:26 +0000)] 
Improve test

6 years agoInterface files now include promotion flags
Simon Peyton Jones [Wed, 30 Jan 2013 08:24:59 +0000 (08:24 +0000)] 
Interface files now include promotion flags

6 years agoTest Trac #7609
Simon Peyton Jones [Wed, 30 Jan 2013 08:24:11 +0000 (08:24 +0000)] 
Test Trac #7609

6 years agoRemove some redundant 'extra cleaning' in T4850
Ian Lynagh [Sat, 26 Jan 2013 15:58:19 +0000 (15:58 +0000)] 
Remove some redundant 'extra cleaning' in T4850

6 years agoT5113 is passing, remove expect_broken, see comment on #7046
Simon Marlow [Tue, 29 Jan 2013 09:59:10 +0000 (09:59 +0000)] 
T5113 is passing, remove expect_broken, see comment on #7046

6 years agoupdate haddock.base figures
Simon Marlow [Tue, 29 Jan 2013 09:55:18 +0000 (09:55 +0000)] 
update haddock.base figures

6 years agoSimpleFail9 now passes, which is actually fine
Simon Peyton Jones [Tue, 29 Jan 2013 08:52:33 +0000 (08:52 +0000)] 
SimpleFail9 now passes, which is actually fine

6 years agoTest Trac #7524
Simon Peyton Jones [Mon, 28 Jan 2013 21:04:23 +0000 (21:04 +0000)] 
Test Trac #7524

6 years agoDebug output wibbles
Simon Peyton Jones [Mon, 28 Jan 2013 18:13:11 +0000 (18:13 +0000)] 
Debug output wibbles

6 years agotest for #2435
Simon Marlow [Mon, 28 Jan 2013 09:28:40 +0000 (09:28 +0000)] 
test for #2435

6 years agoMake T7037 work on both Windows and other platforms
Ian Lynagh [Fri, 25 Jan 2013 23:35:28 +0000 (23:35 +0000)] 
Make T7037 work on both Windows and other platforms

6 years agodynCompileExpr(dyn) fails on Windows as ghc isn't built the dyn way
Ian Lynagh [Fri, 25 Jan 2013 20:55:09 +0000 (20:55 +0000)] 
dynCompileExpr(dyn) fails on Windows as ghc isn't built the dyn way

6 years agoTweak a comment to help vim's syntax highlighting
Ian Lynagh [Fri, 25 Jan 2013 20:35:16 +0000 (20:35 +0000)] 
Tweak a comment to help vim's syntax highlighting

6 years agoFix T7037 on Windows
Ian Lynagh [Fri, 25 Jan 2013 20:17:45 +0000 (20:17 +0000)] 
Fix T7037 on Windows

6 years agoHandles TEST_HC=c:/... on Windows
Ian Lynagh [Fri, 25 Jan 2013 19:11:00 +0000 (19:11 +0000)] 
Handles TEST_HC=c:/... on Windows

6 years agonumrun012(ghci) now passes on Windows
Ian Lynagh [Fri, 25 Jan 2013 16:57:31 +0000 (16:57 +0000)] 
numrun012(ghci) now passes on Windows

I don't know why it started working, but we didn't know why it didn't work
before either.

6 years agoRevert "fix runtests to set LD_LIBRARY_PATH environment variable."
Ian Lynagh [Fri, 25 Jan 2013 22:57:19 +0000 (22:57 +0000)] 
Revert "fix runtests to set LD_LIBRARY_PATH environment variable."

This reverts commit d262089127c54bfe81963628ac70a309f8133492.

We shouldn't need to do this on Linux.