Sat Aug 29 12:23:58 CEST 2009 Matthias Kilian <[email protected]> * Make this test a little more stable On slow machines, ghci can take more than one second to launch, which produces confusing failure output like +cat: 3171.pid: No such file or directory +usage: kill [-s signame | -signum | -signame] { job | pid | pgrp } ... + kill -l [exit_status ...] Fix this by 1) increasing the time before sending a SIGINT from 1 to 2 seconds, 2) running the test program in the background and asking the shell for its pid instead of relying on the program to write the pid file quick enough. Of course you'll still see failures on *very* slow or overloaded machines; that would probably a stderr diff like `-Interrupted'.
New patches:
[Make this test a little more stable Matthias Kilian <[email protected]>**20090829102358 Ignore-this: b6e4f541193b785410af602eeb6816aa On slow machines, ghci can take more than one second to launch, which produces confusing failure output like +cat: 3171.pid: No such file or directory +usage: kill [-s signame | -signum | -signame] { job | pid | pgrp } ... + kill -l [exit_status ...] Fix this by 1) increasing the time before sending a SIGINT from 1 to 2 seconds, 2) running the test program in the background and asking the shell for its pid instead of relying on the program to write the pid file quick enough. Of course you'll still see failures on *very* slow or overloaded machines; that would probably a stderr diff like `-Interrupted'. ] hunk ./tests/ghc-regress/ghci/should_run/Makefile 7 # Test that threadDelay can be interrupted by ^C. 3171: - (sleep 1; kill -INT `cat 3171.pid`) & \ - echo "do System.Posix.getProcessID >>= writeFile \"3171.pid\" . show; Control.Concurrent.threadDelay 3000000; putStrLn \"threadDelay was not interrupted\"" | $(TEST_HC) -v0 --interactive 2>3171.err + echo "do Control.Concurrent.threadDelay 3000000; putStrLn \"threadDelay was not interrupted\"" | \ + $(TEST_HC) -v0 --interactive 2>3171.err & \ + sleep 2; kill -INT $$!; wait Context: [update ghc-pkg tests following the addition of installedPackageId Simon Marlow <[email protected]>**20090825075513 Ignore-this: 11e93fe86f380719a29acfaf7b5356b6 ] [follow change in Cabal: package -> sourcePackageId Simon Marlow <[email protected]>**20090825075134 Ignore-this: a8a6691cebf577efcb788fdb771e6f5a ] [Test Trac #3406 [email protected]**20090825073011 Ignore-this: 41fbfb08f09cfffae469f0b8fd44a54b ] [add extra_clean for T3286 Simon Marlow <[email protected]>**20090824091047 Ignore-this: 45f1ed1ca4d9ddeb6851eb71fcc75da0 ] [TFs: T3423 was missing -XFlexibleInstances Manuel M T Chakravarty <[email protected]>**20090824025100 Ignore-this: 748adedd772e71de8ce6ef2923275c7b ] [TFs: T3220 Manuel M T Chakravarty <[email protected]>**20090824024755 Ignore-this: 3032db152dd0668bb1f2f39d8d50b7f ] [Test Trac #2850 [email protected]**20090821211351 Ignore-this: 55733f9b3a70231c38a8aeda2606dd89 ] [Test Trac #3423 [email protected]**20090821210544 Ignore-this: 6fefeb3dfa3fdb3d263b9c4d420792ff ] [Use the dynamic way if we have a dynamic RTS Ian Lynagh <[email protected]>**20090821153355] [Add a dynamic hello world test, that gets run during validate Ian Lynagh <[email protected]>**20090821153314] [Test Trac #3371 [email protected]**20090821100750 Ignore-this: 2f20fa7f0cc2332f35cb3240638a7589 ] [Test Trac #3437 [email protected]**20090821095347 Ignore-this: e3f80cd609ddf79b2b245e7c848bb005 ] [Track wording changes in error messages [email protected]**20090820162121 Ignore-this: cc430b5e53ee4c284655b0aea40d6ce6 ] [add the test from #3424 Simon Marlow <[email protected]>**20090820145325 Ignore-this: 88bc0b2e548875961b69400fa724f0a ] [follow changes in the base package Simon Marlow <[email protected]>**20090819142958 Ignore-this: cb7b532ad97a6048c487376771ff401f ] [accept output Simon Marlow <[email protected]>**20090819142333 Ignore-this: 9f6d0bebea8626427860ae3295239d8c ] [Update output [email protected]**20090820123151 Ignore-this: 4f4cb42b35ca820a0039406b437b1be2 ] [accept output Simon Marlow <[email protected]>**20090819142418] [add threaded2_qw way, which is threaded2 with +RTS -qw Simon Marlow <[email protected]>**20090819130704 Ignore-this: de00b72ca34d0cce9416455b6af5ac66 ] [TFs: T3418 Manuel M T Chakravarty <[email protected]>**20090820105418 Ignore-this: 6fe5caf8af245d792f7b9e4c9be58d68 ] [TFs: test cases for #2767 & #3208 Manuel M T Chakravarty <[email protected]>**20090820073343 Ignore-this: a9c06c0d0fcc72c0cf59e5c94271005b ] [add test for an illegal C finalizer callback Simon Marlow <[email protected]>**20090819123952 Ignore-this: 4de54a86ba9d6d588ea7d6008ba84737 ] [the remaining part of #1548 is now fixed Simon Marlow <[email protected]>**20090819120141 Ignore-this: 4d90dde10d3987b4fab8b4309a5fb6ed ] [Add a test for trac #3286 Ian Lynagh <[email protected]>**20090819113724] [add test for #3429 Simon Marlow <[email protected]>**20090818140030 Ignore-this: 3ad7efc0274575bb7290ca952533c594 ] [Add a test for #3007 Ian Lynagh <[email protected]>**20090813161943] [Test Trac #3409 [email protected]**20090813161540 Ignore-this: e4093d296c05c4b5532588f2a04c2210 ] [Add a test for #3303: multiline deprecated warnings Ian Lynagh <[email protected]>**20090812185743] [Trac change in Template Haskell errors [email protected]**20090811141150 Ignore-this: 979cdffc2f7cf47db2174314dc0429bd ] [Test Trac #3395 [email protected]**20090810142619 Ignore-this: 112740b5a4cff3c501bd8da0cac1ee42 ] [Test Trac #2395 [email protected]**20090810140953 Ignore-this: 7992f4b5a513c2c96645453308a8035a ] [Test for group being a special_id when TransformListComp is on Max Bolingbroke <[email protected]>**20090717224849 Ignore-this: 850192f55096c6d373b797255aa0b236 ] [Regression test for Trac #2395 Alexander Dunlap <[email protected]>**20090807190019 Ignore-this: a60faf89928344e7c9982646ce0193f9 ] [remove deprecated -#include flag Simon Marlow <[email protected]>**20090805105047 Ignore-this: b9aa8cbf54564941c544ae78ba3cb376 ] [fix tests after RTS tidyup Simon Marlow <[email protected]>**20090803123501 Ignore-this: 3ea78cb5bb05e68c634bcc961c13d9a0 ] [Remove the DiffArray tests Ian Lynagh <[email protected]>**20090802132048 They're now in the diffarray package ] [Add a test for trac #789 Ian Lynagh <[email protected]>**20090801153538] [Update array test outputs Ian Lynagh <[email protected]>**20090730153649] [add test for atomic_inc()/atomic_dec() Simon Marlow <[email protected]>**20090729075547 Ignore-this: f754a4aaa6aa21458375e299f2e7141b ] [2816 is if_platform('i386-unknown-mingw32',expect_broken(3398)) Simon Marlow <[email protected]>**20090727143202 Ignore-this: a9e5f453ce3b5a951c2364cfcebd7cbe ] [bump limits for T1969 again (it was failing on Windows) Simon Marlow <[email protected]>**20090724091355 Ignore-this: b384a8221a8dec7a89522ee79de2dbf6 ] [Test for make supporting abspath, and fail if it doesn't Ian Lynagh <[email protected]>**20090726135136 This fixes problems using GNU make 3.80. ] [add --no-user-package-conf for ghc-pkg tests Simon Marlow <[email protected]>**20090724151549 Ignore-this: 58db107bb49d62173ad46881cd956599 ] [add -no-user-package-conf to GHC command lines Simon Marlow <[email protected]>**20090724123813 Ignore-this: 57b0e90bfcb89d1793a2c61c014b5414 ] [Test Trac #3391 [email protected]**20090723160523 Ignore-this: dbd9e3f6cb4ce6564aa1d6e9c8339a28 ] [Follow output changes (braces in do-notation) [email protected]**20090723152556 Ignore-this: 74e9ad4d85c7624114084ad8eb054910 ] [fix broken test on Windows Simon Marlow <[email protected]>**20090723120826 Ignore-this: b7be929f7d1302f59cd74cf1eff4aade ] [Follow output [email protected]**20090723131845 Ignore-this: 78fe9e7daf9d70c06da6dfb0332575f7 ] [Test standalone deriving for GADTs [email protected]**20090723100441 Ignore-this: b45fafc180241619d364f1080215869f ] [Tweak tests for unboxed tuple sections Max Bolingbroke <[email protected]>**20090717214231 Ignore-this: 6abc5631f1028b628f155f0a20ece4da ] [Tests for unboxed tuple sections Max Bolingbroke <[email protected]>**20090717212443 Ignore-this: 3a31b5c69acb53e3f543c8fb92aca38e ] [Tests for basic TupleSections Max Bolingbroke <[email protected]>**20090717204659 Ignore-this: 7bf3a8e08b6b654f5dd9a9fba31f9cbe ] [Use /usr/bin/env to find Python Simon Marlow <[email protected]>**20090723075156 Ignore-this: e6315057539fb198d98909cb1dda243f Contributed by: Krister Walfridsson <[email protected]> ] [Follow Integer changes Ian Lynagh <[email protected]>**20090721234536] [accept output Simon Marlow <[email protected]>**20090722122529 Ignore-this: 35b847e78b8062dac668425dcf3e7a57 ] [Give us some more breathing room in T1969. Simon Marlow <[email protected]>**20090722122518 Ignore-this: 550cf0b0b58bf69f57858dccc9e4e416 Memory use went up a little with my change to external names, as most names got longer. However, we're still doing better than 6.10 on this test, so I don't mind bumping the limit a bit. ] [tc215 works now that Trac #366 is done [email protected]**20090722105733 Ignore-this: e9f108b8632316cc2d9c23370121d03 ] [Test pattern-match overlap checking for GADTs [email protected]**20090722051806 Ignore-this: 355ff54d49f196f3b4e769ce486786f0 ] [Test for Trac #3382 [email protected]**20090720060155 Ignore-this: b8a90bfdf4219235cf0adb51c0d36e36 ] [Add a test for #1647 Ian Lynagh <[email protected]>**20090719181425] [Add a test for #3055 Ian Lynagh <[email protected]>**20090717222040] [Add test for Trac #3346 [email protected]**20090717155827 Ignore-this: 52203cff9520f0f502612cf5880e660c ] [accept output after changes to dfun naming Simon Marlow <[email protected]>**20090716144125 Ignore-this: 295a3d8c86c533edb40d077a1fbdf2e5 ] [Update mode001 Ian Lynagh <[email protected]>**20090716000721 The earlier of --help and --version now determines the behaviour ] [add utf8_bom codec Simon Marlow <[email protected]>**20090715131505 Ignore-this: 27a198bd9ed3112005c43551109acd58 ] [Test derived Foldable instance [email protected]**20090711130821 Ignore-this: 830f4b824bd469df0de947f32b4f9c1c ] [Add a test for #1074 Ian Lynagh <[email protected]>**20090711191937] [Add a test for #1792 Ian Lynagh <[email protected]>**20090711191508] [TFs: Added T2203b Manuel M T Chakravarty <[email protected]>**20090710064638 Ignore-this: 8d0c3101f03a76b2fd13e1032aefeffe ] [TFs: fixed should_compile/Simple8 & wibbles Manuel M T Chakravarty <[email protected]>**20090710035536 Ignore-this: 66b4c2ad6a5f594f2fd6fcf51be78d6a ] [T3016: skip_if_fast, and omit optc too. Simon Marlow <[email protected]>**20090709144504 Ignore-this: 26d695f84bf0393da3f1e1862140384f ] [disable T3016(profc), gcc takes ages Simon Marlow <[email protected]>**20090709104147 Ignore-this: 4accafb20b11f18bb389be86aa84a331 ] [Tests for unused imports [email protected]**20090706111329 Ignore-this: 477db02945a12e468d0adc9181ece35d ] [TFs: T2677 Manuel M T Chakravarty <[email protected]>**20090707054749 Ignore-this: 8ca0fd55918fc3b5a7cd2b7407b7195a ] [Update cabal01 test for Cabal change Ian Lynagh <[email protected]>**20090705194639] [Make changes to -fwarn-unused-do-bind and -fwarn-wrong-do-bind suggested by SPJ Max Bolingbroke <[email protected]>**20090702150957 Ignore-this: a79f826df7152b7b5a253a05f90d4128 ] [Support for -fwarn-unused-do-bind and -fwarn-wrong-do-bind, as per #3263 Max Bolingbroke <[email protected]>**20090701200441 Ignore-this: b762c27276c3e1e3aff614640f27903d ] [Test Trac #3342 [email protected]**20090702124700 Ignore-this: 3d47a4e0c60d0ad4db266869cdb74ec9 ] [Track error message change [email protected]**20090702095512 Ignore-this: 47db428ab8f9a6e2f0903fc84c1d547b ] [NonLinearSigErr is actually OK [email protected]**20090702095452 Ignore-this: ef3db790608ce2d9b4a26cbc450b93c1 ] [Track change in record syntax for GADTs [email protected]**20090702095341 Ignore-this: f566b1130a4dff0a81d92a262d794d ] [TFs: nullary families are ok Manuel M T Chakravarty <[email protected]>**20090702093629 Ignore-this: ef5783432881e51f4f88b806aaacc1cf ] [add readwrite003 (from bug #679) Simon Marlow <[email protected]>**20090701104449 Ignore-this: 657cea9c9506a5f961877cdb77313ab7 ] [Remove hacky GCC setting Simon Marlow <[email protected]>**20090626140827 Ignore-this: 698a64c4f09be46340d04aeb0f9be8d It isn't necessary now that Cabal can find gcc in the new layout, and it was taking a long time (2.5s on Windows every time you start up make in the testsuite) ] [accept output after fixes to FFI declaration pretty-printing Simon Marlow <[email protected]>**20090626103342 Ignore-this: 34d49ce46f4fac185e110ae6c27e8c35 ] [add test for #3319 Simon Marlow <[email protected]>**20090626103159 Ignore-this: 7b77c0bb4137b9174e76c6c5aa9a9bd4 ] [TAG 2009-06-25 Ian Lynagh <[email protected]>**20090625160429] Patch bundle hash: 593af773c49f2bffb6c148ceae2459632925597e
_______________________________________________ Cvs-ghc mailing list [email protected] http://www.haskell.org/mailman/listinfo/cvs-ghc
