#4915: Segmentation fault or access violation in GHCi (Windows) during tab-
completion
------------------------+---------------------------------------------------
Reporter: drvitek | Owner:
Type: bug | Status: new
Priority: normal | Component: GHCi
Version: 6.12.3 | Keywords:
Testcase: | Blockedby:
Os: Windows | Blocking:
Architecture: x86 | Failure: None/Unknown
------------------------+---------------------------------------------------
While working in GHCi, I tried to tab-complete "readFile" after typing in
"rea". GHCi immediately quit and either GHCi or the system wrote the
following to standard output:
"Segmentation fault/access violation in generated code"
In addition, the recent commands (accessed by pressing up/down) upon
opening a new session of GHCi behaved unusually. In particular, the last
command run before the tab-complete segfault was ":r", which completed
successfully. However, upon opening a new GHCi session, the command
accessed by pressing the Up key was ":t fmap handleInput $ join $ fmap
(sequence . map readFile) getArgs", which was one of the first commands
run in that session of GHCi.
Note that probably between one and two hundred commands had been executed
in the current GHCi session upon failure, and several of those commands
had printed somewhat large strings to standard output (on the order of ten
commands at around 10K chars each).
The specific syntax which set off the segfault occured after typing
"(genelist,opts) <- handleInput (readFile "inputs\\genes.fasta") (rea" and
then pressing Tab.
All attempts to recreate the bug failed. Dump of all accessible GHCi
history is provided. Note that the "fmap handleInput ..." command is not
accessible; I don't have the expertise to access the entire dump, and as
mentioned above I cannot recreate it as the previous command history is
not properly functioning.
Further system notes:
OS: Windows 7 32-bit
Finally, I apologize for this being spectacularly unhelpful in not
providing a test case at all. I tried for several minutes to recreate
some semblance of the environment in which the bug was found;
unfortunately, this bug seems to be more elusive than a mosquito in a dark
room.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4915>
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
