Mail Archives: cygwin/2001/07/06/02:23:33
Unless you know good reasons to do otherwise, the standard cygwin
installation is far better than the usual one size fits all. As we
pointed out, it's not cygwin per se which has trouble with NAI; all gzip
family and many other styles of compressed files produce false
positives.
If you don't understand the prefix systems of gcc and cygwin, you're
better off letting it default to /usr/local. Cygwin builds of gcc 3.x
follow fairly closely the standard advice on the gcc web pages. You
might want to stick with options which have worked for others,
like --enable-sjlj-exceptions (I also use --with-included-gettext, as it
seems to help on my installation, and --enable-languages='c++,f77' as
those (c implied as well) are supported on cygwin. You should copy
/usr/include to the include directory implied by --prefix
(/usr/local/include for default) befor configuring.
I don't see any great attraction in gcc-3.0.1, as it has a few serious
performance regressions compared to 2.95.3, but it is a step forward in
g++. AFAIK 3.0.1 and 3.1 build the same way, and I find 3.1 impressive
with its better performance (particularly with 8- or 16-byte alignment
set in binutils) and beginnings of ability to do -msse for P4 (also
dependent on updating binutils), and far better reliability than 3.0 had
until near release.
----- Original Message -----
From: "ted byers" <ted DOT byers AT sympatico DOT ca>
To: <cygwin AT cygwin DOT com>
Sent: Thursday, July 05, 2001 7:34 PM
Subject: My attempt to build gcc-3.0 failed :-(
> Could you direct me on how to fix the problem I encountered, bearing
in mind
> that I have not built a compiler before, and am entirely new to cygwin
(I am
> using NT4, sp6, if that matters). I also have emacs installed, and
had it
> well before I installed cygwin, and was hoping to figure out how to
> integrate use of the gcc compilers and gnumake into emacs. But it
will take
> a little while to refresh my memory regarding what I used to do on
unix.
> :-(
>
> I installed most of your distribution, possibly more than I needed,
after
> some trepidation. Your remarks about NAI virus software not liking
> something about your install procedure made me overly cautious. But I
> decided to take a chance and downloaded setup from one of your mirrors
and
> ran it, directing it first to just download what I wanted, and then I
ran it
> again to do the installation from a local directory, and that appeared
to
> complete without a problem.
>
> However, I wanted to test gcc-3.0, and you only had gcc-2.95.3 (if I
recall
> correctly), and I remembered seeing something on the gcc3.0 site
indicating
> it should build without a problem using cygwin, so I retrieved the
source
> and tried it. But my attempt failed. Since it does not seem to make
a log
> of the progress made or where and why it failed, I have no idea how
far it
> got into the build process. And as I have never heard of "libtool", I
have
> no idea what it is supposed to do and no idea what the error message
is
> supposed to mean, let alone how to fix it.
>
> I am showing you first the results of mount, so you know where I put
> everything relevant to this effort, and where I want the results put.
The
> Source and Objects directories I am using for gcc3.0 are located in
D:\Gcc.
> I prefer to keep applications I am building as separate as possible
from the
> directory tree containing the applications I am using, so minimize the
> chance that something will happen during the building and testing of
the
> application that might break one of the tools I need.
>
> $ mount
> D:\Gnu\bin on /usr/bin type user (binmode)
> D:\Gnu\lib on /usr/lib type user (binmode)
> D:\Gnu on / type user (binmode)
> c: on /cygdrive/c type user (binmode,noumount)
> d: on /cygdrive/d type user (binmode,noumount)
> e: on /cygdrive/e type user (binmode,noumount)
> f: on /cygdrive/f type user (binmode,noumount)
>
>
>
> Here is the command I used for the comfigure script:
>
> $ sh
> ../Source/gcc-3.0/configure --prefix=d/Gcc --with-local-prefix=d/Gcc -
-enabl
> e-threads=win32 --
> disable-win32-registry
>
>
>
> And here is the command I used to start the build:
> $ make CFLAGS='-O' LIBCFLAGS='-g -O2'
> LIBCXXFLAGS='-g -O2 -fno-implicit-templates' bootstrap-lean
>
>
>
> And here is the mess I found on my screen at the time the build failed
(with
> the errors generated at the end).
>
>
/cygdrive/d/Gcc/Objects/gcc/xgcc -B/cygdrive/d/Gcc/Objects/gcc/ -nostdin
c++
> -L/cygdrive/d/Gcc/Object
>
s/i686-pc-cygwin/libstdc++-v3/src -L/cygdrive/d/Gcc/Objects/i686-pc-cygw
in/l
> ibstdc++-v3/src/.libs -B
> d/Gcc/i686-pc-cygwin/bin/ -Bd/Gcc/i686-pc-cygwin/lib/ -isystem
> d/Gcc/i686-pc-cygwin/include -I../../
>
../../Source/gcc-3.0/libstdc++-v3/../gcc -I../../../../Source/gcc-3.0/li
bstd
> c++-v3/../include -I../.
>
./../../Source/gcc-3.0/libstdc++-v3/include -I../../../../Source/gcc-3.0
/lib
> stdc++-v3/include/std -I
>
../../../../Source/gcc-3.0/libstdc++-v3/include/c_std -I../include -I../
../.
> ./../Source/gcc-3.0/libs
>
tdc++-v3/libsupc++ -g -O2 -fno-implicit-templates -Wall -Wno-format -W -
Wwri
> te-strings -Winline -fdi
> agnostics-show-location=once -ffunction-sections -fdata-sections -g -c
> ../../../../Source/gcc-3.0/li
> bstdc++-v3/libsupc++/vec.cc -o vec.o
> ../../../../Source/gcc-3.0/libstdc++-v3/libsupc++/vec.cc:0:
> warning: -ffunction-sections
> may affect debugging on some targets.
> /bin/sh ../libtool --tag CXX --tag
ared --mode=link
> /cygdrive/d/Gcc/Objects/gcc/
>
gcc -B/cygdrive/d/Gcc/Objects/gcc/ -nostdinc++ -L/cygdrive/d/Gcc/Object
s/i
> 686-pc-cygwin/libstdc++-
>
v3/src -L/cygdrive/d/Gcc/Objects/i686-pc-cygwin/libstdc++-v3/src/.libs -
Bd/G
> cc/i686-pc-cygwin/bin/ -
> Bd/Gcc/i686-pc-cygwin/lib/ -isystem
> c-cygwin/include -fno-implicit-templates
> -Wall -Wno-format -W -Wwrite-strings -Winline -fdiagn
ostics
> -show-location=once
> -ffunction-sections -fdata-sections -g -o
> libsupc++convenience.la del_op.lo del_opnt.l
> o del_opv.lo del_opvnt.lo eh_alloc.lo eh_aux_runtime.lo eh_catch.lo
> eh_exception.lo eh_globals.lo eh
> _personality.lo eh_terminate.lo eh_throw.lo new_handler.lo new_op.lo
> new_opnt.lo new_opv.lo new_opvn
> t.lo pure.lo tinfo.lo tinfo2.lo vec.lo -lm
> mkdir .libs
> ar rc .libs/libsupc++convenience.a del_op.o del_opnt.o del_opv.o
> del_opvnt.o eh_alloc.o eh_aux_runt
> ime.o eh_catch.o eh_exception.o eh_globals.o eh_personality.o
eh_terminate.o
> eh_throw.o new_handler.
> o new_op.o new_opnt.o new_opv.o new_opvnt.o pure.o tinfo.o tinfo2.o
vec.o
> ranlib .libs/libsupc++convenience.a
> creating libsupc++convenience.la
> (cd .libs && rm -f libsupc++convenience.la && ln -s
> ../libsupc++convenience.la libsupc++convenience.
> la)
> /bin/sh ../libtool --tag CXX --tag
ared --mode=link
> /cygdrive/d/Gcc/Objects/gcc/
>
gcc -B/cygdrive/d/Gcc/Objects/gcc/ -nostdinc++ -L/cygdrive/d/Gcc/Object
s/i
> 686-pc-cygwin/libstdc++-
>
v3/src -L/cygdrive/d/Gcc/Objects/i686-pc-cygwin/libstdc++-v3/src/.libs -
Bd/G
> cc/i686-pc-cygwin/bin/ -
> Bd/Gcc/i686-pc-cygwin/lib/ -isystem
> c-cygwin/include -fno-implicit-templates
> -Wall -Wno-format -W -Wwrite-strings -Winline -fdiagn
ostics
> -show-location=once
> -ffunction-sections -fdata-sections -g -o
libsupc++.la -rpath
> d/Gcc/lib del_op.lo del_o
> pnt.lo del_opv.lo del_opvnt.lo eh_alloc.lo eh_aux_runtime.lo
eh_catch.lo
> eh_exception.lo eh_globals.
> lo eh_personality.lo eh_terminate.lo eh_throw.lo new_handler.lo
new_op.lo
> new_opnt.lo new_opv.lo new
> _opvnt.lo pure.lo tinfo.lo tinfo2.lo vec.lo -lm
> libtool: link: only absolute run-paths are allowed
> make[4]: *** [libsupc++.la] Error 1
> make[4]: Leaving directory
> `/cygdrive/d/Gcc/Objects/i686-pc-cygwin/libstdc++-v3/libsupc++'
> make[3]: *** [all-recursive] Error 1
> make[3]: Leaving directory
> `/cygdrive/d/Gcc/Objects/i686-pc-cygwin/libstdc++-v3'
> make[2]: *** [all-recursive-am] Error 2
> make[2]: Leaving directory
> `/cygdrive/d/Gcc/Objects/i686-pc-cygwin/libstdc++-v3'
> make[1]: *** [all-target-libstdc++-v3] Error 2
> make[1]: Leaving directory `/cygdrive/d/Gcc/Objects'
> make: *** [bootstrap-lean] Error 2
>
>
>
> Any assistance you can provide will be greatly appreciated.
>
> Cheers,
>
> Ted
>
> R.E. Byers
> ted DOT byers AT sympatico DOT ca
>
>
> --
> Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting: http://cygwin.com/bugs.html
> Documentation: http://cygwin.com/docs.html
> FAQ: http://cygwin.com/faq/
>
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -