RedHat Linux 6.0 Problems (PR#201)

Martyn Plummer plummer@iarc.fr
Wed, 02 Jun 1999 12:24:43 +0200 (CEST)


On 02-Jun-99 Peter Dalgaard BSA wrote:
> jacobm@magnet.drcmr.dk writes:
> 
>> I had R running in RedHat Linux 5.2. After updating to RedHat Linux 6.0 I
>> get
>> the 
>> following message when I try to start R.
>> /usr/lib/R/bin/R.binary: error in loading shared libraries:
>> /usr/lib/R/bin/R.binary: 
>> undefined symbol: __setfpucw
>> 
>> After that Ive tried to upgrade with the newest RPMS R-0.64.1-4.src.rpm, but
                                                                   ^^^

Did you really rebuild the source rpm, or is this a typo? If the latter, then
this is not an R bug, but a binary incompatibility between RedHat 5.* and 6.0.

> Hmmm. Compiling from raw sources might be easier (and is known to work
> on at least one system!). 
> 
> Anyway, if you're still getting the __setfpucw message when
> rebuilding, then I strongly suspect that the .src.rpm contain the
> results of a run of "configure" on an RH5.x system, resulting in
> 
> /* Some Linux systems may need this */
>#define HAVE___SETFPUCW 1
> 
> in .../src/include/Platform.h. Perhaps it works to remove that line
> and rebuild.

No it doesn't. If you rebuild the source rpm then you run configure.
In fact the build sequence is

./configure
make
make check
make dvi
make install

All the rpm package management stuff is just wrapped around this.
It should not be any different from rebuilding from the pristine source.
In fact I can't understand why the build didn't stop at the make
check stage (There used to be a bug in the Makefile, but I thought this
was fixed now).

I finally have a copy of RedHat 6.0 (although upgrading hasn't gone
too well so far, as usual...) so I should have RH6.0 RPMS out by the
end of the week.

Martyn
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._