The Glasgow Haskell Bugs Mailing List

GHC | 1 Apr 2011 01:22

#5071: GHCi crashes on large alloca/allocaBytes requests

#5071: GHCi crashes on large alloca/allocaBytes requests
---------------------------------+------------------------------------------
    Reporter:  guest             |       Owner:              
        Type:  bug               |      Status:  new         
    Priority:  normal            |   Component:  Compiler    
     Version:  7.0.3             |    Keywords:              
    Testcase:                    |   Blockedby:              
          Os:  Unknown/Multiple  |    Blocking:              
Architecture:  Unknown/Multiple  |     Failure:  None/Unknown
---------------------------------+------------------------------------------
 GHCi, and ghc, seems to exit (crash?) on certain types of memory
 allocation exceptions.

 The FFI addendum says that the allocation functions should be able to
 return null:
 http://www.cse.unsw.edu.au/~chak/haskell/ffi/ffi/ffise5.html#x8-230005

 Section 5.8 says this:

 > If any of the allocation functions fails, a value of Ptr.nullPtr is
 produced. If free or reallocBytes is applied to a memory area that has
 been allocated with alloca or allocaBytes, the behaviour is undefined. Any
 further access to memory areas allocated with alloca or allocaBytes, after
 the computation that was passed to the allocation function has terminated,
 leads to undefined behaviour. Any further access to the memory area
 referenced by a pointer passed to realloc, reallocBytes, or free entails
 undefined behaviour.

 In practice, code examples and documentation appear to rely on `alloca`
 '''NOT''' returning a `nullPtr`:

(Continue reading)

GHC | 1 Apr 2011 01:29

Re: #5071: GHCi crashes on large alloca/allocaBytes requests

#5071: GHCi crashes on large alloca/allocaBytes requests
---------------------------------+------------------------------------------
    Reporter:  guest             |       Owner:              
        Type:  bug               |      Status:  new         
    Priority:  normal            |   Component:  Compiler    
     Version:  7.0.3             |    Keywords:              
    Testcase:                    |   Blockedby:              
          Os:  Unknown/Multiple  |    Blocking:              
Architecture:  Unknown/Multiple  |     Failure:  None/Unknown
---------------------------------+------------------------------------------
Changes (by dmwit):

 * cc: daniel <at> … (added)

--

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5071#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler

_______________________________________________
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs <at> haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs

GHC | 1 Apr 2011 04:46

Re: #5071: GHCi crashes on large alloca/allocaBytes requests

#5071: GHCi crashes on large alloca/allocaBytes requests
---------------------------------+------------------------------------------
    Reporter:  guest             |       Owner:              
        Type:  bug               |      Status:  new         
    Priority:  normal            |   Component:  Compiler    
     Version:  7.0.3             |    Keywords:              
    Testcase:                    |   Blockedby:              
          Os:  Unknown/Multiple  |    Blocking:              
Architecture:  Unknown/Multiple  |     Failure:  None/Unknown
---------------------------------+------------------------------------------

Comment(by guest):

 The Haskell 2010 FFI section agrees that `nullPtr` should be produced:
 http://www.haskell.org/onlinereport/haskell2010/haskellch31.html#x39-28600031

--

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5071#comment:2>

GHC | 1 Apr 2011 04:55

Re: #5071: GHCi crashes on large alloca/allocaBytes requests

#5071: GHCi crashes on large alloca/allocaBytes requests
---------------------------------+------------------------------------------
    Reporter:  guest             |       Owner:              
        Type:  bug               |      Status:  new         
    Priority:  normal            |   Component:  Compiler    
     Version:  7.0.3             |    Keywords:              
    Testcase:                    |   Blockedby:              
          Os:  Unknown/Multiple  |    Blocking:              
Architecture:  Unknown/Multiple  |     Failure:  None/Unknown
---------------------------------+------------------------------------------

Comment(by chak):

 I would suggest to make the libraries comply with the Haskell standard.
 Raising an exception instead of returning a `nullPtr` will break all code
 that is written against the standard.

--

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5071#comment:3>

GHC | 1 Apr 2011 11:07

Re: #5061: Implement FFI spec behaviour for *CString family

#5061: Implement FFI spec behaviour for *CString family
---------------------------------+------------------------------------------
    Reporter:  batterseapower    |        Owner:              
        Type:  bug               |       Status:  patch       
    Priority:  normal            |    Milestone:              
   Component:  Compiler          |      Version:  7.0.3       
    Keywords:                    |     Testcase:              
   Blockedby:                    |   Difficulty:              
          Os:  Unknown/Multiple  |     Blocking:              
Architecture:  Unknown/Multiple  |      Failure:  None/Unknown
---------------------------------+------------------------------------------
Changes (by batterseapower):

  * status:  new => patch

--

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5061#comment:5>

GHC | 1 Apr 2011 11:43

Re: #5052: stdcall on OS X is broken

#5052: stdcall on OS X is broken
-------------------------+--------------------------------------------------
    Reporter:  igloo     |        Owner:              
        Type:  bug       |       Status:  patch       
    Priority:  normal    |    Milestone:              
   Component:  Compiler  |      Version:  7.0.2       
    Keywords:            |     Testcase:              
   Blockedby:            |   Difficulty:              
          Os:  MacOS X   |     Blocking:              
Architecture:  x86       |      Failure:  None/Unknown
-------------------------+--------------------------------------------------
Changes (by batterseapower):

  * status:  new => patch

Comment:

 I've attached a ticket that seems to fix this bug for me. The problem is
 that the codegen for foreign calls makes sure the stack is 16-byte aligned
 on OS X by pushing some padding. In the case where the foreign call is
 cdecl, that padding gets popped after the call, but if the convention is
 stdcall then it doesn't generate any popping code at all. However, this is
 incorrect because the stdcall only promises to pop the arguments, not the
 padding.

 The fix is to generate code to pop the padding (if any) on OS X.

--

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5052#comment:3>

(Continue reading)

GHC | 1 Apr 2011 11:53

Re: #5065: dph tests failing

#5065: dph tests failing
--------------------------------------+-------------------------------------
    Reporter:  igloo                  |        Owner:              
        Type:  bug                    |       Status:  new         
    Priority:  normal                 |    Milestone:              
   Component:  Data Parallel Haskell  |      Version:  7.1         
    Keywords:                         |     Testcase:              
   Blockedby:                         |   Difficulty:              
          Os:  Unknown/Multiple       |     Blocking:  5070        
Architecture:  Unknown/Multiple       |      Failure:  None/Unknown
--------------------------------------+-------------------------------------

Comment(by ezyang):

 The failure I'm getting is:

 {{{
 [1 of 2] Compiling PrimesVect       ( PrimesVect.hs, PrimesVect.o )
 *** Vectorisation error ***
     Variable not vectorised: Control.Exception.Base.patError

 }}}

 So maybe a Core optimization pass isn't managing to eliminate the
 patError?

--

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5065#comment:2>

GHC | 1 Apr 2011 12:09

Re: #4855: Debug.Trace.trace mangles Unicode strings

#4855: Debug.Trace.trace mangles Unicode strings
-------------------------------+--------------------------------------------
    Reporter:  andersk         |        Owner:                             
        Type:  bug             |       Status:  new                        
    Priority:  normal          |    Milestone:  7.4.1                      
   Component:  libraries/base  |      Version:  7.0.1                      
    Keywords:                  |     Testcase:                             
   Blockedby:                  |   Difficulty:                             
          Os:  Linux           |     Blocking:                             
Architecture:  x86_64 (amd64)  |      Failure:  Incorrect result at runtime
-------------------------------+--------------------------------------------

Comment(by batterseapower):

 This happens because GHC call the C function debugBelch to output a
 "trace" string. This has the advantage that the user of GHC can override
 how "trace" messages are handled by changing the debugMsgFn function
 pointer.

 If we were to simply delete the GHC-specific #ifdef in Debug.Trace we
 would be using "hPtStrLn stderr" instead, so the correct locale encoding
 would be applied and trace would handle unicode properly.

 Another option is to fix #5061, at which point we can make Trace work
 properly by using withCString instead of withCAString.

--

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4855#comment:3>

GHC | 1 Apr 2011 12:34

Re: #5065: dph tests failing

#5065: dph tests failing
--------------------------------------+-------------------------------------
    Reporter:  igloo                  |        Owner:              
        Type:  bug                    |       Status:  new         
    Priority:  normal                 |    Milestone:              
   Component:  Data Parallel Haskell  |      Version:  7.1         
    Keywords:                         |     Testcase:              
   Blockedby:                         |   Difficulty:              
          Os:  Unknown/Multiple       |     Blocking:  5070        
Architecture:  Unknown/Multiple       |      Failure:  None/Unknown
--------------------------------------+-------------------------------------
Changes (by simonpj):

 * cc: chak <at> …, benl <at> … (added)

Comment:

 I think Manuel is in mid-flight on some of this.

--

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5065#comment:3>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler

_______________________________________________
Glasgow-haskell-bugs mailing list
Glasgow-haskell-bugs <at> haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs

GHC | 1 Apr 2011 12:37

Re: #5059: Pragma to SPECIALISE on value arguments

#5059: Pragma to SPECIALISE on value arguments
---------------------------------+------------------------------------------
    Reporter:  batterseapower    |        Owner:              
        Type:  feature request   |       Status:  new         
    Priority:  normal            |    Milestone:              
   Component:  Compiler          |      Version:  7.0.3       
    Keywords:                    |     Testcase:              
   Blockedby:                    |   Difficulty:              
          Os:  Unknown/Multiple  |     Blocking:              
Architecture:  Unknown/Multiple  |      Failure:  None/Unknown
---------------------------------+------------------------------------------

Comment(by simonpj):

 Right.  Like I said `inline` would need fixing to be able to inline even
 recursive functions.  But then it might work just fine, right?

--

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5059#comment:5>

Group

gmane.comp.lang.haskell.glasgow.bugs.

Search Archive

Page

1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 61

Articles in period: 609

April 2011

Language

Change language

Options

Current view: Threads only / Showing only 20 lines / Not hiding cited text.
Change to All messages, whole messages, or hide cited text.

Post a message
NNTP Newsgroup
Classic Gmane web interface
XML

XML

RSS Feed
List Information

About Gmane

Recent entries

#7919: Heap corruption (segfault) from large 'let' expression (17 May 2013)
#7918: SrcSpan's associated with expanded quasi-quotes are inconsisten (17 May 2013)
#7917: update documentation of InstalledPackageInfo (17 May 2013)
#7916: PolyKinds without type signatures (16 May 2013)
#7915: Documentation uses deprecated record GADT syntax (16 May 2013)
#781: GHCi on x86_64, cannot link to static data in shared libs (16 May 2013)
#7914: base library's MD5 symbols clash with others (16 May 2013)
#7913: Argument order not preserved by nubBy (15 May 2013)
#7912: Missing symbols from termios.h when linking on Android (15 May 2013)
#7911: The following code makes ghc panic (15 May 2013)
#7910: ConstraintKinds and reifyInstances (15 May 2013)
#7909: GHC doesn't respect -optl and -optc in a couple of places (14 May 2013)
#7908: InstanceSigs suggestion not accepted (14 May 2013)
#910: --make should have a -j flag for parallel building (13 May 2013)
#7907: patch to make comment about stgclosure struct more accurate (13 May 2013)
#7906: INLINEABLE on class method triggers the error for INLINE (11 May 2013)
#7905: Type checker barfs on nonsensical expression (11 May 2013)
#7904: GHC calls the cc1 of a gcc thats not the first in PATH (11 May 2013)
#7903: Failure to find class instance specified with extra parentheses (10 May 2013)
#7902: Add support for byte endianness swapping exposed as a primops (10 May 2013)
[GHC] #7901: Bad error message when using UnicodeSyntax ∀ without Ex (9 May 2013)

Archives

256	May 2013
738	April 2013
472	March 2013
516	February 2013
891	January 2013
359	December 2012
547	November 2012
676	October 2012
412	September 2012
554	August 2012
483	July 2012
2050	June 2012
364	May 2012
546	April 2012
505	March 2012
454	February 2012
654	January 2012
457	December 2011
650	November 2011
432	October 2011
535	September 2011
509	August 2011
609	July 2011
461	June 2011
539	May 2011
602	April 2011
649	March 2011
522	February 2011
378	January 2011
597	December 2010
722	November 2010
813	October 2010
577	September 2010
310	August 2010
488	July 2010
457	June 2010
483	May 2010
419	April 2010
240	March 2010
263	February 2010
415	January 2010
453	December 2009
566	November 2009
352	October 2009
358	September 2009
346	August 2009
423	July 2009
289	June 2009
353	May 2009
755	April 2009
455	March 2009
384	February 2009
428	January 2009
339	December 2008
449	November 2008
543	October 2008
380	September 2008
432	August 2008
538	July 2008
504	June 2008
306	May 2008
364	April 2008
279	March 2008
321	February 2008
395	January 2008
305	December 2007
990	November 2007
390	October 2007
431	September 2007
386	August 2007
505	July 2007
345	June 2007
419	May 2007
211	April 2007
344	March 2007
246	February 2007
363	January 2007
143	December 2006
285	November 2006
375	October 2006
200	September 2006
186	August 2006
112	July 2006
141	June 2006
128	May 2006
80	April 2006
157	March 2006
95	February 2006
193	January 2006
170	December 2005
72	November 2005
88	October 2005
99	September 2005
117	August 2005
147	July 2005
77	June 2005
129	May 2005
154	April 2005
151	March 2005
90	February 2005
81	January 2005
86	December 2004
47	November 2004
62	October 2004
83	September 2004
57	August 2004
47	July 2004
65	June 2004
91	May 2004
38	April 2004
59	March 2004
60	February 2004
43	January 2004
56	December 2003
53	November 2003
82	October 2003
80	September 2003
66	August 2003
88	July 2003
100	June 2003
83	May 2003
86	April 2003
90	March 2003
80	February 2003
77	January 2003
40	December 2002
63	November 2002
58	October 2002
118	September 2002
42	August 2002
42	July 2002

Design

Zawodny | Right Menu | Left Menu

Your Own Design

Paste the URL of your CSS below. Download CSS template