Re: [RFC] Is inside_entry_file useful and needed?

[Corinna Vinschen <vinschen@redhat.com>]

On Thu, Mar 27, 2003 at 12:08:17PM -0500, Andrew Cagney wrote:
> >Hi,
> >
> >the reason I'm going to discuss the inside_entry_file() function is that
> >I had trouble with the asm-source.exp tests on both platforms I currently
> >care about, Cygwin and XStormy16, one native, one embedded target.
> 
> Please note that these tests work on i386 GNU/Linux (and d10v).

I don't doubt that they work on other targets but they apparently fail
due to the calls of the inside_entry_file() for these two targets.

Note that Linux doesn't need the inside_entry_file() function.  All
tests are still working as expected if all calls are dropped as in
my patch on gdb-patches.

> >frame.c::get_prev_frame():
> >[...]
> This thest terminates the unwind after there has been a stack frame in 
> the startup file.  Remember, the d10v, which only relies on this test, 
> works.

Please note that I don't say to remove the inside_entry_file() tests
entirely from gdb.  But it's obviously wrong to call this stuff inside
of the generic functionality.  This should remain in some way inside
of the target specific code.  Either by having a way to disable the calls
for a target or by having a architecture method which is called from
inside the inside_entry_file() function itself.

> Is there something significnatly different between i386 GNU/Linux and, 
> say, cygwin?

I'm not quite sure what you're up to.  There are a whole bunch of
differences probably, e. g. how applications (even assembler apps)
are linked.

Corinna

-- 
Corinna Vinschen
Cygwin Developer
Red Hat, Inc.
mailto:vinschen@redhat.com