#4462: -dcore-lint error in simplifier phase 0 when profiling
-------------------------------------------------+--------------------------
Reporter: wkahl | Owner:
Type: bug | Status: new
Priority: normal | Milestone:
Component: Compiler | Version: 7.0.1 RC1
Keywords: profiling, simplifier, core-lint | Testcase: Agda (darcs,
current)
Blockedby: | Difficulty:
Os: Linux | Blocking:
Architecture: x86_64 (amd64) | Failure: Compile-time
crash
-------------------------------------------------+--------------------------
Comment(by wkahl):
> GHC should never segfault unless you are using unsafeCoerce or the FFI.
Building Agda apparently includes compiling
{{{
dist/build/Agda/Syntax/Parser/Parser.hs
}}}
which seems to be generated by the Agda compiler and therefore consists
largely of {{{unsafeCoerce}}}, but I could not find where it is called,
and in any case, by Agda's behaviour I do not think it is parsing while it
crashes on me.
Agda works somewhat like {{{ghc --make}}}, and on the first attempt to
typecheck something complex I almost certainly get a crash, but not
necessarily on the second attempt, which uses the {{{.agdai}}} files the
first attempt created. I have not yet started to try to delete these for
reproduction, but I am currently concentrating on crashes that reproduce
directly, and preferably with more information than just a segfault. With
the latter aspect, I haven't been successful with GHC-7 yet, but I have a
reproducible segfault that apparently still is a segfault (second attempt
running right now) with the binary compiled with {{{-debug}}}.
With ghc-6.12.3 I have been more successful, and I just submitted
ticket:4472 so that you can at least see that.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4462#comment:5>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
