Re: [Rd] i may have missed something ..

2007-01-27 Thread Jan de Leeuw
verified -- jan On Jan 27, 2007, at 14:01 , Simon Urbanek wrote: > > On Jan 26, 2007, at 11:22 AM, Jan de Leeuw wrote: > >> Switching the icc compiler flag from -O3 to -O0 for deriv.c solves >> the problem. As I said, I have to do that for regex.c as well. >> > > Just for the record - the latte

Re: [Rd] i may have missed something ..

2007-01-27 Thread Simon Urbanek
On Jan 26, 2007, at 11:22 AM, Jan de Leeuw wrote: > Switching the icc compiler flag from -O3 to -O0 for deriv.c solves > the problem. As I said, I have to do that for regex.c as well. > Just for the record - the latter is due to a bug in the regex code: one of the functions is declared "pure

Re: [Rd] i may have missed something ..

2007-01-26 Thread Jan de Leeuw
Switching the icc compiler flag from -O3 to -O0 for deriv.c solves the problem. As I said, I have to do that for regex.c as well. -- Jan On Jan 26, 2007, at 07:52 , Prof Brian Ripley wrote: > Ouch. It does look like a compiler over-optimization sort of problem. > > I presume that is the ix86 ic

Re: [Rd] i may have missed something ..

2007-01-26 Thread Jan de Leeuw
Yes. No such problem with the pair gcc/ifort. I'll try less aggressive optimization with icc on deriv.c (I also have to do that for regex.c). -- Jan On Jan 26, 2007, at 07:52 , Prof Brian Ripley wrote: > Ouch. It does look like a compiler over-optimization sort of problem. > > I presume that is

Re: [Rd] i may have missed something ..

2007-01-26 Thread Prof Brian Ripley
Ouch. It does look like a compiler over-optimization sort of problem. I presume that is the ix86 icc, with which we have not had much success on either Linux or Windows. I've just checked x86_64 icc on Linux, and that is working correctly. Brian On Fri, 26 Jan 2007, Jan de Leeuw wrote: > Thi

Re: [Rd] i may have missed something ..

2007-01-26 Thread Jan de Leeuw
This is R revision 40576 compiled with icc/ifort on OS X 10.4.9 (8P2122). It may be the compiler. -- J. > example(deriv) deriv> ## formula argument : deriv> dx2x <- deriv(~ x^2, "x") ; dx2x expression({ .value <- x^2 .grad <- array(0, c(length(.value), 1), list(NULL, c("x"))) .gr

Re: [Rd] i may have missed something ..

2007-01-26 Thread Prof Brian Ripley
On Thu, 25 Jan 2007, Jan de Leeuw wrote: > but deriv() and friends do not work in R-devel (at least > not on the Mac). They work for me under Linux and Windows. What does example(deriv) give you? > == > Jan de Leeuw, 11667 Steinhoff Rd, F

[Rd] i may have missed something ..

2007-01-26 Thread Jan de Leeuw
but deriv() and friends do not work in R-devel (at least not on the Mac). == Jan de Leeuw, 11667 Steinhoff Rd, Frazier Park, CA 93225 home 661-245-1725 skype 661-347-0667 global 254-381-4905 .mac: jdeleeuw +++ aim: deleeuwjan +++ skype: j_de