[GHC] #14825: Access violation on Windows is not propagated to the OS

GHC ghc-devs at haskell.org
Tue Feb 20 13:28:57 UTC 2018


#14825: Access violation on Windows is not propagated to the OS
-------------------------------------+-------------------------------------
        Reporter:  varosi            |                Owner:  (none)
            Type:  feature request   |               Status:  closed
        Priority:  normal            |            Milestone:
       Component:  Runtime System    |              Version:  8.2.2
      Resolution:  duplicate         |             Keywords:
Operating System:  Windows           |         Architecture:  x86_64
 Type of failure:  Debugging         |  (amd64)
  information is incorrect           |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #13911            |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by Phyx-):

 Replying to [comment:3 varosi]:
 >
 > Is it possible to give some stack trace in such situation instead just
 "Segmentation fault/access violation in generated code"? I mean even if
 exact throwing place is not known, but only partial stack trace
 information. Or you mean that in GHC 8.4 partial stack information will be
 possible?

 yup, GHC 8.4 will give an attempt, see
 https://phabricator.haskell.org/D3913

 If you're running the code in GHCi I have information from the runtime's
 symbol table to help out, but when running compiled I don't have this.
 Eventually (when I get some time) I want to add the ability to use STABS
 information if available. That should help a bit.

 Also the stack tracer in GHC 8.4 cannot unwind between Haskell and C
 stacks. Since I don't know where I am when the segfault occurs I don't
 know which way to unwind (should be possible though). So for now I assume
 it's C code.

 GHC 8.4 also supports creating crash dumps
 https://phabricator.haskell.org/D3912
 So you can just attach a debugger later and inspect the dump manually.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14825#comment:4>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list