[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
20000229: Error with Make All on Mcidas 7.6 at Colgate
- Subject: 20000229: Error with Make All on Mcidas 7.6 at Colgate
- Date: Tue, 29 Feb 2000 10:47:41 -0700
>From: Adam Burnett <address@hidden>
>Organization: Colgate
>Keywords: 200002291726.KAA26578 McIDAS-X 7.60
Adam,
>It's been a long time since I've harassed you with Mcidas questions.
Yes, it has!
>I was
>just getting around to building mcidas 7.6 (better late than never)
Boy, I'll say. FYI, I will be attempting to get the next McIDAS distribution
ready for release by the end of July.
>when I
>discovered a small problem. Things move along nicely until I get the
>following error:
>
>DNO_UNION_WAIT=1 -DRETSIGTYPE=void -DHAVE_SIGNED_CHAR=1 -DHAVE_SYS_IOCTL_H=1
>-DHAVE_SYS_FILIO_H=1 -DTCL_SHLIB_EXT=\".so\"
>./../generic/tclExecute.c
>"./../generic/tclExecute.c", line 328: undefined symbol: hypot
>"./../generic/tclExecute.c", line 328: non-constant initializer: op "NAME"
>"./../generic/tclExecute.c", line 413: cannot recover from previous errors
>c89: acomp failed for ./../generic/tclExecute.c
>*** Error code 2
>make: Fatal error: Command failed for target `tclExecute.o'
>Current working directory /data2/mcidas/mcidas7.6/tcl/unix
>*** Error code 1
>make: Fatal error: Command failed for target `.built_tcl'
>gissun%
OK, the problem is in the Tcl/Tk portion of the build. I suggest doing
the following:
define CC to be cc instead of c89:
setenv CC cc
cd mcidas7.6/tcl/unix
make distclean
cd ../../tk/unix
make distclean
cd ../../src
rm tclcomp
make all
>As some background information, my compilers are Solaris 5.0. I poked
>around on the web support documents but couldn't find anything like this.
Hopefully, you did read the warnings about SC5.0 in the Notes and Warnings
page: http://www.unidata.ucar.edu/packages/mcidas/mcx/warnings_mcx.html.
>Any thoughts? Do I have an environmental variable messed up somewhere?
I would guess that the problem is related to use of c89 instead of cc,
but the error seems to relate to the math library (by virtue of the
undefined symbol: hypot).
>Thanks
Please let me know if doing the above doesn't get you going.
Tom