Re: [Rd] R crash with complex matrix algebra when using EISPACK=TRUE

From: Prof Brian Ripley <ripley_at_stats.ox.ac.uk>
Date: Tue 03 Jan 2006 - 11:56:18 GMT

Well, valgrind helps a lot more here: it points at lines 2131 and 2135-6 of eigen.f, and using gdb shows those are called with NA = 0.

So it is a bug in EISPACK, and easy enough to fix -- now done.

On Tue, 3 Jan 2006, Ole F. Christensen wrote:

> Brain, Thank you very much for your help.
> Using gctorture I was able to produce a simple function call showing the
> problem.
>
>
> Gm <- rbind(c(-0.3194373786, 0.2444066686, 0.0428108831, 3.221983e-02),
> c(0.0002071301, -0.0003282719, 0.0001211418,
> 5.128830e-12),
> c(0.0621332005, 0.0545850010, -0.2098487035,
> 9.313050e-02),
> c(0.0280936142, 0.0586642184, 0.1658310277,
> -2.525889e-01)
> )
> print(Gm)
> temp <- eigen(Gm)
> print(temp)
> gctorture(TRUE)
> temp <- eigen(Gm, EISPACK = TRUE)
>
>
>
> ###
> # On my computer I get :
> ##
>
> > gctorture(TRUE)
> >
> > source("http://www.daimi.au.dk/~olefc/TEST/Gm.R")
> > print(Gm)
> [,1] [,2] [,3] [,4]
> [1,] -0.3194373786 0.2444066686 0.0428108831 3.221983e-02
> [2,] 0.0002071301 -0.0003282719 0.0001211418 5.128830e-12
> [3,] 0.0621332005 0.0545850010 -0.2098487035 9.313050e-02
> [4,] 0.0280936142 0.0586642184 0.1658310277 -2.525889e-01
> >
> > temp <- eigen(Gm)
> > print(temp)
> $values
> [1] -3.464342e-01+1.3161e-03i -3.464342e-01-1.3161e-03i
> [3] -8.933476e-02+0.0000e+00i 9.052031e-19+0.0000e+00i
>
> $vectors
> [,1] [,2] [,3]
> [1,] -0.3419128709-0.03748199i -0.3419128709+0.03748199i -0.222056433+0i
> [2,] 0.0003508445+0.00001921i 0.0003508445-0.00001921i 0.001421758+0i
> [3,] -0.4179745717+0.01301547i -0.4179745717-0.01301547i -0.664932225+0i
> [4,] 0.8407249376+0.00000000i 0.8407249376+0.00000000i -0.713129761+0i
> [,4]
> [1,] -0.5+0i
> [2,] -0.5+0i
> [3,] -0.5+0i
> [4,] -0.5+0i
>
> > temp <- eigen(Gm, EISPACK = TRUE)
> *** glibc detected *** free(): invalid pointer: 0x08c4e778 ***
> Aborted
>
>
>
> # Should I submit this as a bug report also ?
>
>
> Best
>
> Ole
>
>
>
> Prof Brian Ripley wrote:
>
>> Try valgrind. That is reporting use outside arrays in rg, that is the
>> non-complex case of eigen().
>>
>> Otherwise, using gctorture(TRUE) will help precipitate the error.
>>
>> On Mon, 2 Jan 2006, Ole F. Christensen wrote:
>>
>>>
>>> Dear subscribers of R-devel
>>>
>>> I am experiencing that R crashes (further details are given below) in
>>> some complex matrix calculations when EISPACK=TRUE has been specified in
>>> eigen().
>>> I discovered the behaviour some months ago just after the
>>> release of R-2.2.0, and it has been lying on my desk since.
>>> I apologise for not having nailed the problem down to a simple function
>>> call, but I thought I should better report the problem now
>>> instead of waiting.
>>> My hope is that someone will either spot the cause
>>> of the crash in a minute, or otherwise provide some help for me to
>>> investigate further.
>>>
>>>
>>> ## The code :
>>>
>>>
>>> source("http://www.daimi.au.dk/~olefc/TEST/fct.R")
>>> source("http://www.daimi.au.dk/~olefc/TEST/parm.crash.R")
>>>
>>> for(l.v in 1:4){
>>> for(r.v in 1:4){
>>> for(l.x in 1:4){
>>> for(r.x in 1:4){
>>> hvad <-
>>> inhomoWmat.complex(subst.ratematrix(parm.same.str.sym$Gamma[,,l.v,r.v]),subst.ratematrix(parm.same.str.sym$Gamma[,,l.x,r.x]),
>>>
>>> EISPACK=TRUE)
>>> print(c(l.v,r.v,l.x,r.x))
>>> }
>>> }
>>> }
>>> }
>>>
>>>
>>> ## gives
>>>
>>> [1] 1 1 1 1
>>> [1] 1 1 1 2
>>> [1] 1 1 1 3
>>> [1] 1 1 1 4
>>> [1] 1 1 2 1
>>> Segmentation fault
>>>
>>>
>>>
>>> ### whereas the code seems to work fine when EISPACK=FALSE :
>>>
>>>
>>> source("http://www.daimi.au.dk/~olefc/TEST/fct.R")
>>> source("http://www.daimi.au.dk/~olefc/TEST/parm.crash.R")
>>>
>>> for(l.v in 1:4){
>>> for(r.v in 1:4){
>>> for(l.x in 1:4){
>>> for(r.x in 1:4){
>>> hvad <-
>>> inhomoWmat.complex(subst.ratematrix(parm.same.str.sym$Gamma[,,l.v,r.v]),subst.ratematrix(parm.same.str.sym$Gamma[,,l.x,r.x]),
>>>
>>> EISPACK=FALSE)
>>> print(c(l.v,r.v,l.x,r.x))
>>> }
>>> }
>>> }
>>> }
>>>
>>> ## works fine.
>>>
>>>
>>> ## There is some randomness in how and when the crash happens.
>>> ## The crash is either :
>>>
>>> Segmentation fault
>>>
>>> #
>>>
>>> *** glibc detected *** double free or corruption (!prev): 0x08aa7298
>>> ***
>>>
>>> # or
>>>
>>> *** glibc detected *** free(): invalid pointer: 0x082bfd20 ***
>>>
>>>
>>>
>>> ## Seen on R-2.2-1 and R-2.2-0 .
>>> ## Not seen in R-2.1.1 !
>>> ## I haven't investiated whether it happens on Windows also.
>>>
>>>
>>> ### A few details on the matrix calculations :
>>> The eigenvalue decomposition is done on 4 * 4 matrices where the rows
>>> sum to 0.
>>> The matrices may be on the edge of not being complex diagonalizable.
>>>
>>>
>>> version
>>> _
>>> platform i686-pc-linux-gnu
>>> arch i686
>>> os linux-gnu
>>> system i686, linux-gnu
>>> status
>>> major 2
>>> minor 2.1
>>> year 2005
>>> month 12
>>> day 20
>>> svn rev 36812
>>> language R
>>>
>>>
>>> Thanks in advance of any help.
>>>
>>> Ole Christensen
>>>
>>> --
>>> Ole F. Christensen
>>> BiRC - Bioinformatics Research Center
>>> University of Aarhus
>>>
>>> ______________________________________________
>>> R-devel@r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-devel
>>>
>>>
>>
>
> --
> Ole F. Christensen
> BiRC - Bioinformatics Research Center
> University of Aarhus
>
> ______________________________________________
> R-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
>

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

______________________________________________
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Received on Tue Jan 03 23:17:13 2006

This archive was generated by hypermail 2.1.8 : Tue 03 Jan 2006 - 14:26:30 GMT