public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
* directory names in linker
@ 1998-03-21 13:50 Ruslan Shevchenko
  1998-03-24 13:00 ` Jim Wilson
  0 siblings, 1 reply; 5+ messages in thread
From: Ruslan Shevchenko @ 1998-03-21 13:50 UTC (permalink / raw)
  To: egcs

After installing egcs-2.91.14 and compiling programs with
-belph, I receive error:
crtbegin.o: can not open file for reading

warning: /usr/local/lib/gcc-lib/elph/egcs-2.91.14/ not used
(or some in this style, I don't remember)

after
ln -s /usr/local/lib/gcc-lib/i586-pc-sco3.2v5.0.4/egcs-2.91.14
     /usr/local/lib/gcc-lib/elf/egcs-2.91.14

all work.


--

    @=
     //RSSH                              mailto:Ruslan@Shevchenko.Kiev.UA




^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: directory names in linker
  1998-03-21 13:50 directory names in linker Ruslan Shevchenko
@ 1998-03-24 13:00 ` Jim Wilson
  1998-03-24 14:49   ` Ruslan Shevchenko
  0 siblings, 1 reply; 5+ messages in thread
From: Jim Wilson @ 1998-03-24 13:00 UTC (permalink / raw)
  To: rssh; +Cc: egcs

	After installing egcs-2.91.14 and compiling programs with
	-belph, I receive error:
	crtbegin.o: can not open file for reading

-b is a gcc option that does not do what you think it does.  See the gcc
documentation.

Try using -melf/-mcoff instead.  I believe -melf is the default, so you
shouldn't need to specify it.

Jim

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: directory names in linker
  1998-03-24 13:00 ` Jim Wilson
@ 1998-03-24 14:49   ` Ruslan Shevchenko
  1998-03-26 21:49     ` Jim Wilson
  0 siblings, 1 reply; 5+ messages in thread
From: Ruslan Shevchenko @ 1998-03-24 14:49 UTC (permalink / raw)
  To: Jim Wilson; +Cc: rssh, egcs

Jim Wilson wrote:

>         After installing egcs-2.91.14 and compiling programs with
>         -belph, I receive error:
>         crtbegin.o: can not open file for reading
>
> -b is a gcc option that does not do what you think it does.  See the gcc
> documentation.
>
> Try using -melf/-mcoff instead.  I believe -melf is the default, so you
> shouldn't need to specify it.
>

I think egcs must have backward compability with gcc.for example, -b,elph
is from Makefile for sun Tcl-8.0

Of course, we can say, that it is bug in sun configure,
but if they checked it with gcc, why it is not must work
with egcs ?


> Jim



--

    @=
     //RSSH                              mailto:Ruslan@Shevchenko.Kiev.UA




^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: directory names in linker
  1998-03-24 14:49   ` Ruslan Shevchenko
@ 1998-03-26 21:49     ` Jim Wilson
  1998-03-29  5:14       ` -belf " Robert Lipe
  0 siblings, 1 reply; 5+ messages in thread
From: Jim Wilson @ 1998-03-26 21:49 UTC (permalink / raw)
  To: rssh; +Cc: egcs

I don't know what you are talking about, but -belf has never done what you
think it does, and this has nothing to do with EGCS vs gcc.  Please see
the gcc documentation.

-melf is the correct option for both EGCS and gcc.

It would be possible to build a copy of gcc where -belf did something useful,
but that would require doing a lot of stuff by hand.  Perhaps someone once
distributed a binary package for gcc that was hand constructed in such a way
that -belf worked.  However, you can't expect -belf to work for a compiler that
you build yourself unless do the same hackery.  Since you have not done so,
in your build, the -belf option won't work for you.

Or perhaps you are confusing SCO C compiler options with gcc options.  Maybe
-belf if the correct SCO C compiler option?

Jim


^ permalink raw reply	[flat|nested] 5+ messages in thread

* -belf Re: directory names in linker
  1998-03-26 21:49     ` Jim Wilson
@ 1998-03-29  5:14       ` Robert Lipe
  0 siblings, 0 replies; 5+ messages in thread
From: Robert Lipe @ 1998-03-29  5:14 UTC (permalink / raw)
  To: Jim Wilson; +Cc: egcs

> I don't know what you are talking about, but -belf has never done what you
> think it does, and this has nothing to do with EGCS vs gcc.  Please see
> the gcc documentation.

I was mystified, too.   I was going to pipe up, but since he didn't bother
to name the host or target, I didn't bother to comment.

> It would be possible to build a copy of gcc where -belf did something useful,
> but that would require doing a lot of stuff by hand.  Perhaps someone once
> distributed a binary package for gcc that was hand constructed in such a way
> that -belf worked.  However, you can't expect -belf to work for a compiler that

For SCO OpenServer, some dude did indeed distribute a wildly popular (a couple
of tens of thousands distributed via download and CD) version of gcc that did
support '-belf', just like the OpenServer native compiler did.

I can't remember who that dude was, but I have it on Good Authority that
it was actually based on a Cygnus progressive tree becuase that's what
the dude happened to have. :-)

In fact, for this target, it's even explictly mentioned in the EGCS doc:
	http://egcs-www.cygnus.com/install/specific.html
with some additional info in the SCO prgrammer FAQ at
	http://www.dgii.com/cgi-bin/scoprgfaq/faq.pl

RJL

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~1998-03-29  5:14 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1998-03-21 13:50 directory names in linker Ruslan Shevchenko
1998-03-24 13:00 ` Jim Wilson
1998-03-24 14:49   ` Ruslan Shevchenko
1998-03-26 21:49     ` Jim Wilson
1998-03-29  5:14       ` -belf " Robert Lipe

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).