Test failures with -DDEBUG stage2

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Test failures with -DDEBUG stage2

GHC - devs mailing list

Devs

Should I expect failures with a stage2 compiler built with -DDEBUG?   I’d much prefer the answer to be “no”.

But I several failure

Unexpected failures:

   deriving/should_compile/T11148.run         T11148 [bad exit code (2)] (normal)

   indexed-types/should_compile/T11361.run    T11361 [exit code non-0] (normal)

   polykinds/T11362.run                       T11362 [exit code non-0] (normal)

   stranal/should_compile/T9208.run           T9208 [exit code non-0] (optasm)

   typecheck/should_fail/TcCoercibleFail.run  TcCoercibleFail [stderr mismatch] (normal)

 

Fragile tests:

   libraries/base/tests/CPUTime001.run   CPUTime001 [fragile pass] (normal)

   profiling/should_run/heapprof001.run  heapprof001 [fragile fail] (ghci-ext-prof)

   profiling/should_run/heapprof001.run  heapprof001 [fragile fail] (prof)

   profiling/should_run/T15897.run       T15897 [fragile fail] (profasm)

   ghci/should_run/T3171.run             T3171 [fragile pass] (normal)

   profiling/should_run/T5559.run        T5559 [fragile fail] (ghci-ext-prof)

   profiling/should_run/T5559.run        T5559 [fragile fail] (prof)

   concurrent/should_run/T5611.run       T5611 [fragile pass] (normal)

   concurrent/should_run/T5611a.run      T5611a [fragile pass] (normal)

 

Of these, the first three are

ghc-stage2: compiler/cbits/genSym.c:14: checkUniqueRange: Assertion `u != UNIQUE_MASK' failed.

What’s going on there?

Simon


_______________________________________________
ghc-devs mailing list
[hidden email]
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs
Reply | Threaded
Open this post in threaded view
|

Re: Test failures with -DDEBUG stage2

Ben Gamari-2
Simon Peyton Jones via ghc-devs <[hidden email]> writes:

> Devs
> Should I expect failures with a stage2 compiler built with -DDEBUG?
> I'd much prefer the answer to be "no".

The answer is indeed "no" and I have been working on fixing these in
!1296 but I've been reluctant to commit until all tests are green.
I'll try to get a subset of my fixes merged soon to at least resolve the
issues that you are seeing.

Cheers,

- Ben


_______________________________________________
ghc-devs mailing list
[hidden email]
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

signature.asc (497 bytes) Download Attachment