id summary reporter owner description type status priority milestone component version resolution keywords cc os architecture failure difficulty testcase blockedby blocking related 5090 print026 sporadically fails ezyang igloo "It seems to still be the case that print026 sporadically fails: {{{ =====> print026(ghci) 25 of 74 [0, 0, 0] cd . && HC='/home/ezyang/Dev/ghc-build-master/inplace/bin/ghc-stage2' HC_OPTS='-dcore-lint -dcmm-lint -dno-debug-output -no-user-package-conf -rtsopts ' '/home/ezyang/Dev/ghc-build-master/inplace/bin/ghc-stage2' --interactive -v0 -ignore-dot-ghci -dcore-lint -dcmm-lint -dno-debug-output -no-user-package-conf -rtsopts -ignore-dot-ghci print026.run.stdout 2>print026.run.stderr Actual stdout output differs from expected: --- ./print026.stdout.normalised 2011-04-04 16:58:58.000000000 +0100 +++ ./print026.run.stdout.normalised 2011-04-04 16:58:58.000000000 +0100 @@ -1,7 +1,4 @@ fromList ""abc"" -l = Data.Sequence.Seq (Data.Sequence.Deep 3 - (Data.Sequence.One (Data.Sequence.Elem 'a')) - Data.Sequence.Empty - (Data.Sequence.Two (Data.Sequence.Elem 'b') - (Data.Sequence.Elem 'c'))) +l = ( 'a') + ( 'b' 'c') l = fromList ""abc"" *** unexpected failure for print026(ghci) }}} I saw this commit by Ian a while back: {{{ commit 5e95d99276e754a12279b909ab6365c262e80da1 Author: Ian Lynagh Date: Mon Jun 2 20:41:13 2008 +0000 Skip print026 in the 6.8 branch It fails for me, but not consistently - possibly only with a validate build. Seems plausible that the problem is something like library build flags. -l = Data.Sequence.Seq (Data.Sequence.Deep 3 - (Data.Sequence.One (Data.Sequence.Elem 'a')) Data.Sequence.Empty - (Data.Sequence.Two (Data.Sequence.Elem 'b') (Data.Sequence.Elem 'c'))) +l = ( 'a') + ( 'b' 'c') }}} So this is probably not some deep bug, but we ought fix up the test case to work in all cases. (I can't tell if this is a containers problem or a GHC problem.)" bug closed normal 7.2.1 Compiler 7.1 worksforme Unknown/Multiple Unknown/Multiple None/Unknown