fast test suite failures on OS X with 7.10 RC head

hey all heres the results from the test suite run on a build of 7.10 head on OS X i believe that some of the test suite failures are because I set the min OS X version to 10.6 rather than 10.7, and that some are because GCC (unlike clang) doesn't support certain OS X specific things like objective C or related header files. I'll add more information as I have the time, but figured I should share this for now Unexpected results from: TEST="T10019 sigcabal01 objc-hi objcpp-hi retc001 linker_unload debug T2276_ghci T9203 T1969 T9675 T783 T5642 T6048" OVERALL SUMMARY for test run started at Sun Mar 22 14:58:07 2015 EDT 0:31:34 spent to go through 4405 total tests, which gave rise to 18027 test cases, of which 13966 were skipped 52 had missing libraries 3950 expected passes 45 expected failures 0 caused framework failures 0 unexpected passes 8 unexpected failures 6 unexpected stat failures Unexpected failures: cabal/sigcabal01 sigcabal01 [bad stderr] (normal) codeGen/should_compile debug [bad stderr] (normal) driver/objc objc-hi [exit code non-0] (normal) driver/objc objcpp-hi [exit code non-0] (normal) driver/retc001 retc001 [bad stderr] (normal) ffi/should_run T2276_ghci [bad exit code] (ghci) rts linker_unload [bad exit code] (normal) th T10019 [bad stdout] (ghci) Unexpected stat failures: perf/compiler T1969 [stat not good enough] (normal) perf/compiler T5642 [stat not good enough] (normal) perf/compiler T6048 [stat not good enough] (optasm) perf/compiler T783 [stat not good enough] (normal) perf/compiler T9675 [stat not good enough] (optasm) perf/should_run T9203 [stat too good] (normal) ~/D/r/g/testsuite (ghc-7.10|…) $ echo $MACOSX_DEPLOYMENT_TARGET 10.6

Maybe it's time for a Mac OS X taskforce?
All these unexpected test failures on OS X (see also
https://ghc.haskell.org/trac/ghc/ticket/9389), and OS X specific
tickets in general, could get some more attention:
https://ghc.haskell.org/trac/ghc/query?status=!closed&os=MacOS+X
-Thomas (who, like most ghc developers it seems, doesn't work on a Mac)
On Sun, Mar 22, 2015 at 8:40 PM, Carter Schonwald
hey all
heres the results from the test suite run on a build of 7.10 head on OS X
i believe that some of the test suite failures are because I set the min OS X version to 10.6 rather than 10.7, and that some are because GCC (unlike clang) doesn't support certain OS X specific things like objective C or related header files. I'll add more information as I have the time, but figured I should share this for now
Unexpected results from: TEST="T10019 sigcabal01 objc-hi objcpp-hi retc001 linker_unload debug T2276_ghci T9203 T1969 T9675 T783 T5642 T6048"
OVERALL SUMMARY for test run started at Sun Mar 22 14:58:07 2015 EDT 0:31:34 spent to go through 4405 total tests, which gave rise to 18027 test cases, of which 13966 were skipped
52 had missing libraries 3950 expected passes 45 expected failures
0 caused framework failures 0 unexpected passes 8 unexpected failures 6 unexpected stat failures
Unexpected failures: cabal/sigcabal01 sigcabal01 [bad stderr] (normal) codeGen/should_compile debug [bad stderr] (normal) driver/objc objc-hi [exit code non-0] (normal) driver/objc objcpp-hi [exit code non-0] (normal) driver/retc001 retc001 [bad stderr] (normal) ffi/should_run T2276_ghci [bad exit code] (ghci) rts linker_unload [bad exit code] (normal) th T10019 [bad stdout] (ghci)
Unexpected stat failures: perf/compiler T1969 [stat not good enough] (normal) perf/compiler T5642 [stat not good enough] (normal) perf/compiler T6048 [stat not good enough] (optasm) perf/compiler T783 [stat not good enough] (normal) perf/compiler T9675 [stat not good enough] (optasm) perf/should_run T9203 [stat too good] (normal)
~/D/r/g/testsuite (ghc-7.10|…) $ echo $MACOSX_DEPLOYMENT_TARGET 10.6
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

I work exclusively on Mac and have been seeing those errors for months. I'll admit to not having looked into them, as I was fairly sure they Weren't My Fault. However, I'd be willing to be a supportive member of a Mac taskforce, reporting what I see and running experiments as requested. I'm afraid I can't volunteer to be a primary member of such a taskforce, as I know very little about the sorts of things that cause platform-specific failures.
Do let me know if anyone out there wants more information (dumps, other experiments) related to the failures above.
Thanks!
Richard
On Mar 22, 2015, at 4:24 PM, Thomas Miedema
Maybe it's time for a Mac OS X taskforce?
All these unexpected test failures on OS X (see also https://ghc.haskell.org/trac/ghc/ticket/9389), and OS X specific tickets in general, could get some more attention: https://ghc.haskell.org/trac/ghc/query?status=!closed&os=MacOS+X
-Thomas (who, like most ghc developers it seems, doesn't work on a Mac)
On Sun, Mar 22, 2015 at 8:40 PM, Carter Schonwald
wrote: hey all
heres the results from the test suite run on a build of 7.10 head on OS X
i believe that some of the test suite failures are because I set the min OS X version to 10.6 rather than 10.7, and that some are because GCC (unlike clang) doesn't support certain OS X specific things like objective C or related header files. I'll add more information as I have the time, but figured I should share this for now
Unexpected results from: TEST="T10019 sigcabal01 objc-hi objcpp-hi retc001 linker_unload debug T2276_ghci T9203 T1969 T9675 T783 T5642 T6048"
OVERALL SUMMARY for test run started at Sun Mar 22 14:58:07 2015 EDT 0:31:34 spent to go through 4405 total tests, which gave rise to 18027 test cases, of which 13966 were skipped
52 had missing libraries 3950 expected passes 45 expected failures
0 caused framework failures 0 unexpected passes 8 unexpected failures 6 unexpected stat failures
Unexpected failures: cabal/sigcabal01 sigcabal01 [bad stderr] (normal) codeGen/should_compile debug [bad stderr] (normal) driver/objc objc-hi [exit code non-0] (normal) driver/objc objcpp-hi [exit code non-0] (normal) driver/retc001 retc001 [bad stderr] (normal) ffi/should_run T2276_ghci [bad exit code] (ghci) rts linker_unload [bad exit code] (normal) th T10019 [bad stdout] (ghci)
Unexpected stat failures: perf/compiler T1969 [stat not good enough] (normal) perf/compiler T5642 [stat not good enough] (normal) perf/compiler T6048 [stat not good enough] (optasm) perf/compiler T783 [stat not good enough] (normal) perf/compiler T9675 [stat not good enough] (optasm) perf/should_run T9203 [stat too good] (normal)
~/D/r/g/testsuite (ghc-7.10|…) $ echo $MACOSX_DEPLOYMENT_TARGET 10.6
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs
participants (3)
-
Carter Schonwald
-
Richard Eisenberg
-
Thomas Miedema