validate fails!

[Simon Peyton-Jones]

See previous messages re simpl019 and TH_runio.  I have no idea about divbyzero and outofmem.

S

From: cvs-ghc-bounces at haskell.org [mailto:cvs-ghc-bounces at haskell.org] On Behalf Of naur at post11.tele.dk
Sent: 11 October 2007 06:54
To: Manuel M T Chakravarty
Cc: cvs-ghc at haskell.org
Subject: Re: validate fails!

Hello,

On my PPC Mac OS X, validate fails with:

Unexpected passes:
   simpl019(normal)

Unexpected failures:
   TH_runIO(normal)
   divbyzero(normal)
   outofmem(normal)

(But at least I was able to complete the validation and getting to the testsuite, not least thanks to your fixing the additional warnings in ghci/InteractiveUI.hs.)

divbyzero has been failing for a while, but it is not clear what to do about it: It is a matter of the C implementation that allows division by zero to give a zero result.

Som additional details:

=====> TH_runIO(normal)
cd ./th && '/Users/thorkilnaur/tn/GHCDarcsRepository/ghc-HEAD-complete-for-pulling-and-copying-20070713_1212/ghc/compiler/stage2/ghc-inplace' -no-recomp -dcore-lint -dcmm-lint -Dpowerpc_apple_darwin  -c TH_runIO.hs  -v0 -fth -package template-haskell >TH_runIO.comp.stderr 2>&1
Actual stderr output differs from expected:
--- ./th/TH_runIO.stderr.normalised     2007-10-11 03:56:54.000000000 +0200
+++ ./th/TH_runIO.comp.stderr.normalised        2007-10-11 03:56:54.000000000 +0200
@@ -6,5 +6,5 @@
              fail = fail @ IO $dMonad
              $dMonad = GHC.IOBase.$f16
            in runIO @ Exp (fail @ Exp "hi")
-    In the expression: $[splice](runIO (fail "hi"))
-    In the definition of `foo': foo = $[splice](runIO (fail "hi"))
+    In the expression: $(runIO (fail "hi"))
+    In the definition of `foo': foo = $(runIO (fail "hi"))
*** unexpected failure for TH_runIO(normal)

So the TH_runIO problem could be as simple as a changed manner of reporting this error.

And finally:

=====> outofmem(normal)
cd ./rts && '/Users/thorkilnaur/tn/GHCDarcsRepository/ghc-HEAD-complete-for-pulling-and-copying-20070713_1212/ghc/compiler/stage2/ghc-inplace' -no-recomp -dcore-lint -dcmm-lint -Dpowerpc_apple_darwin  -o outofmem outofmem.hs    >outofmem.comp.stderr 2>&1
cd ./rts && ./outofmem    </dev/null >outofmem.run.stdout 2>outofmem.run.stderr
Actual stderr output differs from expected:
--- ./rts/outofmem.stderr.normalised    2007-10-11 03:50:25.000000000 +0200
+++ ./rts/outofmem.run.stderr.normalised        2007-10-11 03:50:25.000000000 +0200
@@ -1 +1 @@
-outofmem: out of memory
+outofmem: memory allocation failed (requested 1074790400 bytes)
*** unexpected failure for outofmem(normal)

Again a case of a changed error report.

Thanks and best regards
Thorkil

Manuel M T Chakravarty <chak at cse.unsw.edu.au> wrote:
The testsuite run of validate failed this morning with

Unexpected passes:
simpl019(normal)

Unexpected failures:
TH_runIO(normal)

simpl019 has just not been marked as passing after #1746 was fixed.
I'll push a patch to the testsuite changing that shortly.

However, I have no idea what the issue with TH_runIO is.

Manuel

_______________________________________________
Cvs-ghc mailing list
Cvs-ghc at haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.haskell.org/pipermail/cvs-ghc/attachments/20071011/446f35f9/attachment.htm