On 19 Jul 2001, Dag-Erling Smorgrav wrote:

> Top of tree:
> 
> --------------------------------------------------------------
> >>> stage 4: building libraries
> --------------------------------------------------------------
> ...
> cd /usr/src/gnu/lib/csu; make _EXTRADEPEND
> cc -O -pipe -march=k6 -I/usr/src/gnu/lib/csu/../../../contrib/gcc.295/config -I. 
>-DIN_GCC  -finhibit-size-directive -fno-inline-functions  -fno-exceptions 
>-fno-omit-frame-pointer -I/sec/obj/usr/src/i386/usr/include  -g0 -DCRT_BEGIN  -c -o 
>crtbegin.o  /usr/src/gnu/lib/csu/../../../contrib/gcc.295/crtstuff.c 
>/usr/src/gnu/lib/csu/crtbegin.s
> cc: cannot specify -o with -c or -S and multiple compilations
> *** Error code 1

I saw this too.  It is caused by garbage in the source tree (crtbegin.s
in this case) and bug(s) in in the rule to create crtbegin.o).  I'm
not sure what created the garbage.  For me, I think it was caused by
playing with cross-compiles (I tried MACHINE_ARCH=alpha and
MACHINE_ARCH=pc98 last week).

> I'd also like to know why it took *six hours* to get there on a 350
> MHz K6-2 with 512 MB RAM.

Maybe just creeping features, but I think they only account for half of
the slowdown since 3.x.

Bruce


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Reply via email to