Hi again,
That gets it past the acconfig step. Analagously, I received messages
like:
Make: Don't know how to make ./macros/../sigc++/basic_signal.h.m4. Stop.
In this case, I had to create a directory sigc++ in the existing sigc++
directory. I then touched the files:
adaptor.h.m4 handle.h.m4 scope.h.m4 thread.h.m4
basic_signal.h.m4 marshal.h.m4 sigc++config.h.m4 type.h.m4
bind.h.m4 object.h.m4 slot.h.m4
Then I touched the *.h versions in the sigc++ directory. That got me past
the remaining missing files messages!
I do have a new problem though, but I think it relates to the gcc
compiler. At least I had this problem when I originally compiled the gcc
compiler:
Making all in sigc++
make all-recursive
Making all in macros
No suffix list.
/bin/sh ./libtool --mode=compile c++ -DHAVE_CONFIG_H -I. -I. -I. -I./..
-I./.. -g -O2 -c adaptor.cc
c++ -DHAVE_CONFIG_H -I. -I. -I. -I./.. -I./.. -g -O2 -c adaptor.cc
echo timestamp > adaptor.lo
/bin/sh ./libtool --mode=compile c++ -DHAVE_CONFIG_H -I. -I. -I. -I./..
-I./.. -g -O2 -c thread.cc
c++ -DHAVE_CONFIG_H -I. -I. -I. -I./.. -I./.. -g -O2 -c thread.cc
mips-tfile, /tmp/ccGyOQtW.s:8 Invalid .stabs/.stabn directive, value not
found
line: #.stabs "/usr/local/src/lyx-1.1.6fix1/sigc++/",100,0,0,$Ltext0
mips-tfile, /tmp/ccGyOQtW.s:9 Invalid .stabs/.stabn directive, value not
found
line: #.stabs "thread.cc",100,0,0,$Ltext0
*** Exit 1
Stop.
*** Exit 1
Stop.
*** Exit 1
Stop.
*** Exit 1
Stop.
The people at Compaq suggested that there was a bug in the tru64 assembler
and I had to create the following script:
#!/bin/sh
/usr/bin/as -old $*
which simply calls as with the -old argument. This worked to compile the
gcc compiler, but is not working here ... :(
Any other suggestions?
Dave
On Thu, 26 Apr 2001, Allan Rae wrote:
> On Wed, 25 Apr 2001, David Bailey wrote:
>
> > Hi Allan,
> >
> > Actually, there is an acconfig.h in the sigc++ directory already. It is
> > identical to the one you sent out.
> >
> > It sounds like something more peculiar to the make utility on the tru64
> > operating system?
>
> Hmmm... It would seem that either there is a timestamp problem and make
> wants to regenerate acconfig.h because it's timestamp is older than
> [what??]. In which case `touch acconfig.h` might fix the problem as I
> suggested at the previous occasion. Does this have any effect?
>
> _Or_ what I'm now thinking might be the case:
> The %.h rule in the Makefile causes make to try to find something to
> compare every header files time stamp against. If you
>
> touch macros/acconfig.h.m4 ; sleep 2 ; touch acconfig.h
>
> does this help?
>
> If it does then it'll probably stop at whatever next header file doesn't
> have a corresponding macros/%h.m4 file. Hmmm... strange if that is the
> case.
>
> Please let me know what happens and then we can inform the libsigc++
> maintainer with a detailed report.
>
> Allan. (ARRae)
>
David A. Bailey, [EMAIL PROTECTED]
School of Oceanography, University of Washington
Box 357940, Seattle, WA 98195-7940 USA
ph: +1 206 221 6570 fax: +1 206 685 3354
