|
[ ghc-Bugs-1198393 ] Program aborts with segmentation fault when profiling : msg#00009lang.haskell.glasgow.bugs
Bugs item #1198393, was opened at 2005-05-09 16:21 Message generated for change (Settings changed) made by simonmar You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108032&aid=1198393&group_id=8032 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: Profiling Group: None >Status: Closed >Resolution: Works For Me Priority: 5 Submitted By: Nobody/Anonymous (nobody) Assigned to: Nobody/Anonymous (nobody) Summary: Program aborts with segmentation fault when profiling used. Initial Comment: I have a program that I have compiled under ghc6.5 (downloaded from the CVS repository on May 5/6), on Fedora-2 core running on a PC. This will run fine when compiled without profiling. With profiling enabled (-prof -auto-all with GHC, and +RTS -p for the program), it aborts, with the message "internal error: Segmentation fault". The same program will compile and run, both with and without profiling, under ghc6.2.1. I have also tried with different datasets, and the same pattern occurs. I updated my ghc installation from CVS today (9th May), but have not been able to update my ghc installation to see if this problem has been removed, as the fptools-level build process is failing, with the Haddock build complaining that parsec-1.0 is an unknown package when building Utils.hs; building in the ghc dir results in a failure to find a module Version when building Main.hs in the utils/ghc-pkg area. The program and dataset are contained in the attached gzip'd tar file. I compile it (without profiling) using ghc --make -fglasgow-exts -o Reader Reader.hs ---------------------------------------------------------------------- >Comment By: Simon Marlow (simonmar) Date: 2005-06-03 09:33 Message: Logged In: YES user_id=48280 Closing: it worked for me, and the submitter has not followed up. ---------------------------------------------------------------------- Comment By: Simon Marlow (simonmar) Date: 2005-05-11 10:06 Message: Logged In: YES user_id=48280 I have not been able to reproduce this one, either with 6.4 or with HEAD of today (11/05). Can you verify that it now works for you? ---------------------------------------------------------------------- Comment By: Nobody/Anonymous (nobody) Date: 2005-05-09 16:23 Message: Logged In: NO Submitter forgot to add: contact email address is David Duke, email djd@xxxxxxxxxxxxxxxx ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108032&aid=1198393&group_id=8032
|
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| Previous by Date: | RE: Behaviour of +RTS -Hx -Mx -RTS options on win32, Simon Marlow |
|---|---|
| Next by Date: | [ ghc-Bugs-1209129 ] program seqfaults when translated with profiling, SourceForge.net |
| Previous by Thread: | [ ghc-Bugs-1213876 ] Xlib Types Not Instances of Anything, SourceForge.net |
| Next by Thread: | [ ghc-Bugs-1209129 ] program seqfaults when translated with profiling, SourceForge.net |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
| News | FAQ | advertise |