On Mon, May 31, 2010 at 10:56:51PM +1000, tony@develop-help.com wrote: > Automated smoke report for 5.13.1 patch 1408fb84de0c28007d91730cde177d893e427181 v5.13.1-114-g1408fb8 > mars: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz (GenuineIntel 1596MHz) (x86_64/4 cpu) > on linux - 2.6.26-2-amd64 [debian] > using icc version 10.1 20080801 > smoketime 3 hours 15 minutes (average 16 minutes 17 seconds) > Failures: (common-args) -Dcc=icc > [stdio/perlio/de_DE.utf8] > [stdio/perlio/de_DE.utf8] -DDEBUGGING > [stdio/perlio/de_DE.utf8] -Dusemorebits > [stdio/perlio/de_DE.utf8] -DDEBUGGING -Dusemorebits > [stdio/perlio/de_DE.utf8] -Dmad > [stdio/perlio/de_DE.utf8] -DDEBUGGING -Dmad > [stdio/perlio/de_DE.utf8] -Duseithreads > [stdio/perlio/de_DE.utf8] -DDEBUGGING -Duseithreads > [stdio/perlio/de_DE.utf8] -Duseithreads -Dusemorebits > [stdio/perlio/de_DE.utf8] -DDEBUGGING -Duseithreads -Dusemorebits > [stdio/perlio/de_DE.utf8] -Duseithreads -Dmad > [stdio/perlio/de_DE.utf8] -DDEBUGGING -Duseithreads -Dmad > ../t/op/magic.t.............................................FAILED > 84-85 Do you have the full output from this test? I'm very confused, as it fails these two on Win32 (but inconsistently between George Greer's smoker, and Steve Hays), and now here with icc on Linux, and I have no idea why. Nicholas ClarkThread Next