validate HEAD on win/xp: CPUTime001, ThreadDelay001

[Claus Reinke]

if i apply Duncan's fix for bytestring, and restrict myself to
building only the things required for validate, two of validate's
tests still fail "unexpected"ly for me:

    OVERALL SUMMARY for test run started at Mon Oct 22 02:00:58 GMTDT 2007
        1971 total tests, which gave rise to
        7517 test cases, of which
           0 caused framework failures
        5875 were skipped

        1564 expected passes
          76 expected failures
           0 unexpected passes
           2 unexpected failures

    Unexpected failures:
       CPUTime001(normal)
       ThreadDelay001(normal)

ThreadDelay001 is fairly persistent, and i reported its output
here earlier; CPUTime001 appears more randomly, probably
depending on system load:

=====> CPUTime001(normal)
cd ./lib/CPUTime && 
'c:/fptools/ghc/compiler/stage2/ghc-inplace' -no-recomp -dcore-lint -dcmm-lint -Di386_unknown_mingw32 
  -o CPUTime001 CPUTime001.hs    >CPUTime001.comp.stderr 2>&1
cd ./lib/CPUTime && ./CPUTime001    </dev/null >CPUTime001.run.stdout 2>CPUTime001.run.stderr
Actual stdout output differs from expected:
--- ./lib/CPUTime/CPUTime001.stdout.normalised 2007-10-22 02:08:26.593750000 +0100
+++ ./lib/CPUTime/CPUTime001.run.stdout.normalised 2007-10-22 02:08:26.593750000 +0100
@@ -1,4 +1,4 @@
 1028457
 1664079
 2692537
-(True,True)
+(False,True)
*** unexpected failure for CPUTime001(normal)

claus