Re: Hide __[u]{div,mod}di3

[Franz Sirl <Franz dot Sirl-kernel at lauterbach dot com>]

On Monday 27 January 2003 22:36, Andreas Schwab wrote:
> Roland McGrath <roland@redhat.com> writes:
> |> > Franz Sirl <Franz.Sirl-kernel@lauterbach.com> writes:
> |> > |> On Sunday 26 January 2003 01:47, Andreas Schwab wrote:
> |> > |> > +compat_symbol (libc, __umoddi3, __umoddi3, GLIBC_2_0);
> |> > |>
> |> > |> Shouldn't that be GLIBC_2.0 (. not _)?
> |> >
> |> > No, this must be a valid identifier, which is translated via
> |> > abi-versions.h into the actual version (which might not be GLIBC_2.0
> |> > after all).  It's a wrapper around symbol_version.
> |> >
> |> > |> I also wonder why you didn't wrap it with a SHLIB_COMPAT?
> |> >
> |> > I'm not sure whether it is needed.
> |>
> |> If you don't use SHLIB_COMPAT, you will compile in unexported dead code
> |> on a configuration that doesn't need to support GLIBC_2.0.
>
> These functions will never be dead, since they are used by glibc itself.

That's clear, but the runtime reference @GLIBC_2.0 should go away at some 
point in the future if requested.

Franz.