No, that was all together in one patch, 5 days ago:
https://github.com/ghc/ghc/commit/7ed482d909556c1b969185921e27e3fe30c2fe86
On Wed, Nov 26, 2014 at 1:37 PM, Simon Peyton Jones
Oh, if you did that in the last day or two, I might have missed it. I’ll try again
*From:* josepedromagalhaes@gmail.com [mailto:josepedromagalhaes@gmail.com] *On Behalf Of *José Pedro Magalhães *Sent:* 26 November 2014 11:27 *To:* Simon Peyton Jones *Cc:* ghc-devs@haskell.org *Subject:* Re: T5462Yes1
I thought I had solved that by removing the extra-clean for GFunctor.o from another test, but apparently the file is still being deleted at some point. Note that T56462Yes1, T5462Yes2, and T5462No all use GFunctor:
test('T5462Yes1', normal, multimod_compile_and_run, ['T5462Yes1', '-iGEq -iGEnum -iGFunctor'])
test('T5462Yes2', normal, multimod_compile_and_run, ['T5462Yes2', '-iGFunctor'])
test('T5462No1', normal, multimod_compile_fail, ['T5462No1', '-iGFunctor'])
On Wed, Nov 26, 2014 at 12:22 PM, Simon Peyton Jones < simonpj@microsoft.com> wrote:
I’m getting this on Linux:
Unexpected failures:
generics T5462Yes1 [exit code non-0] (normal)
Linking T5462Yes1 ...
gcc: error: GEq/GEq1A.o: No such file or directory
gcc: error: GFunctor/GFunctor.o: No such file or directory
*** unexpected failure for T5462Yes1(normal)
But it doesn’t happen if I say “make TEST=T5462Yes1”.
Could this be a multi-thread file deletion thing? make TEST=T5462Yes1 runs the test 8 different ways
Simon
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs