[R-SIG-Mac] quantile() bus error

[urb_r at b-q-c.com]

Jennifer,

you didn't tell us what tools you use, but my guess would be that  
you're using Apple's gcc which is broken. Although I don't quite see  
the point in building R 2.3.1 from sources on a Mac, you can do so  
using gcc 4.0.3 that comes with the R 2.3.1 binary or is available  
separately from
http://r.research.att.com
Alternatively you should be able to use any recent gcc >= 4.0.3 as  
long as you have a matching gfortran and Apple's gcc driver.

Cheers,
Simon

Quoting Jennifer Small <jenlsmall at verizon.net>:

> Hello list,
>
> I'm attempting to build R 2.3.1 from source on a Mac Mini intel duo
> core with OS 10.4.7.  There were no errors reported in configure or
> make. However, make check fails immediately on the quantile function.
> The trace is
>
>  > x <- sort(rnorm(100));  cH <- 1.35
>  > pmin(cH, quantile(x)) # no names
>
> *** caught bus error ***
> address 0x107bcb1, cause 'non-existent physical address'
>
> Traceback:
> 1: sort(x, partial = unique(c(lo, hi)))
> 2: quantile.default(x)
> 3: quantile(x)
> 4: pmin(cH, quantile(x))
> aborting ...
>
> If I open an R command line session and try to call quantile() I get
> the same bus error:
>
>  > quantile(1:10)
>
> *** caught bus error ***
> address 0x107bcb1, cause 'non-existent physical address'
>
> Traceback:
> 1: sort(x, partial = unique(c(lo, hi)))
> 2: quantile.default(1:10)
> 3: quantile(1:10)
>
> Possible actions:
> 1: abort (with core dump)
> 2: normal R exit
> 3: exit R without saving workspace
> 4: exit R saving workspace
> Selection: q
> Selection: 1
> aborting ...
> Bus error
>
> I'm stumped. Any ideas? Thanks in advance.
>
> _______________________________________________
> R-SIG-Mac mailing list
> R-SIG-Mac at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
>
>
>