Re: >Re: make error on redhat 6.2and celeron 400

[msokolov at ivan dot Harhan dot ORG (Michael Sokolov)]

rosely@yeah.net wrote:

> I download gcc-2.95.2.tar.gz and binutils-2.9.tar.gz.Then I configure and make binutils and gcc.

binutils-2.9 are obsolete. Use binutils-2.10 instead. I also strongly recommend
configuring building binutils, gcc, and newlib all together as described in the
CrossGCC FAQ.

> If I use i386-elf for target,make gcc give me following error:configure i386-pc-elf not supported

What do you mean by "make gcc give me following error"? When do you get this
message? Since it's a configure error message, I'm assuming you get it when
running configure for gcc-2.95.2, right? If so it looks like gcc-2.95.2 doesn't
support i386-*-elf. That would be strange, but I can't comment further since
I'm no expert here: I never touched the i386 architecture myself, I work mostly
on vax, m68k, mips, and a few others. If you are indeed getting this message
from gcc-2.95.2's configure script, you should check if this target is
supported in the current CVS gcc, and if not, ask on the gcc mailing list why.
It would seem to me that the elf configuration would be popular with the folks
here, but again, I can't comment on things that involve the i386 architecture,
as I don't do i386.

> If I use i386-aout for target,make gcc give me following error:stdlib.h and unistd.h not such file or direction.

Try configuring and building binutils, gcc, and newlib all together as
described in the CrossGCC FAQ, that is likely to fix this.

> If I use i386-coff for target,make gcc give me following error:toplev.c undefined reference to "sdbout_init".....

This sounds like a gcc configuration bug. (There is one format for -g debug
info that is specific to COFF and is often referred to as the SDB format.
Again, I can't comment on i386, but that's what the m68k-coff configuration
uses by default for example. It looks like in your case gcc is trying to call
the code for doing SDB debug info, but that code isn't built.) Try this with
the current CVS gcc, and if the problem is still there, report the bug.

--
Michael Sokolov		Harhan Engineering Laboratory
Public Service Agent	International Free Computing Task Force
			International Engineering and Science Task Force
			615 N GOOD LATIMER EXPY STE #4
			DALLAS TX 75204-5852 USA

Phone: +1-214-824-7693 (Harhan Eng Lab office)
E-mail: msokolov@ivan.Harhan.ORG (ARPA TCP/SMTP) (UUCP coming soon)