validate test failure

[Simon Peyton-Jones]

David

Any comments?  The patch to interactiveUI was yours I think.  I have not binary-chopped to be sure this is the offending commit but it looks plausible.  And the test is failing consistently

Simon

commit a20cdb930a6f9e980521e0767867217ed96033f4
Author: David Terei <davidterei at gmail.com>
Date:   Wed Dec 21 14:58:39 2011 -0800

    Fix safe imports to work in GHCi.

 compiler/main/GHC.hs     |   26 +++++++++++++--------
 compiler/main/HscMain.hs |   54 +++++++++++++++++++++++++++++++--------------
 ghc/InteractiveUI.hs     |   23 ++++++++++++++-----
 3 files changed, 70 insertions(+), 33 deletions(-)

| -----Original Message-----
| From: cvs-ghc-bounces at haskell.org [mailto:cvs-ghc-bounces at haskell.org] On
| Behalf Of Dimitrios Vytiniotis
| Sent: 22 December 2011 12:19
| To: cvs-ghc at haskell.org
| Subject: validate test failure
| 
| 
| Hi, in my tree I get  'break012' failing with some fairly small modifications
| in the output:
| 
| cd ./ghci.debugger/scripts && HC='/64playpen/dimitris/GHC-
| HEAD/ghc/inplace/bin/ghc-stage2' HC_OPTS='-dcore-lint -dcmm-lint -dno-debug-
| output -no-user-package-conf -rtsopts  -fno-ghci-history '
| '/64playpen/dimitris/GHC-HEAD/ghc/inplace/bin/ghc-stage2' --interactive -v0 -
| ignore-dot-ghci -dcore-lint -dcmm-lint -dno-debug-output -no-user-package-
| conf -rtsopts  -fno-ghci-history  -ignore-dot-ghci   <break012.script
| >break012.run.stdout 2>break012.run.stderr
| Actual stdout output differs from expected:
| --- ./ghci.debugger/scripts/break012.stdout     2011-11-29 12:00:03.000000000
| +0000
| +++ ./ghci.debugger/scripts/break012.run.stdout 2011-12-22 12:17:23.218853401
| +0000
| @@ -6,9 +6,9 @@
|  b :: a2 -> a2 = _
|  c :: () = _
|  d :: a -> a -> a = _
| -a :: t
| -b :: a2 -> a2
| -c :: ()
| +a  :: t
| +b  :: a2 -> a2
| +c  :: ()
|  d :: a -> a -> a
|  a = (_t1::t)
|  b = (_t2::forall a2. a2 -> a2)
| *** unexpected failure for break012(ghci)
| 
| 
| It's not from my latest patch (just pushed) (I checked with HEAD before
| pushing) but it's fairly
| recent (>= 19 Dec). Maybe the latest patches to InteractiveUI etc? Not sure
| ...
| 
| Thanks!
| d-
| 
| 
| _______________________________________________
| Cvs-ghc mailing list
| Cvs-ghc at haskell.org
| http://www.haskell.org/mailman/listinfo/cvs-ghc