Re: [PATCH] Fix PR target/16344

[Geoff Keating <geoffk at geoffk dot org>]

On 05/07/2004, at 11:55 AM, John David Anglin wrote:

> > You do understand that you are patching the symptoms of the bug, not
> > fixing the actual bug itself?  The actual bug is that the PCH file is
> > being built and used with different compiler binaries, probably 
> > because
> > of the long-standing problem that libstdc++'s Makefile doesn't say 
> > that
> > the PCH file depends on the compiler executable, and therefore doesn't
> > rebuild it when the compiler is changed.
>
> Yes, although I consider the SIGSEGV a bug.

Right, but it's not much less of a bug if it doesn't SIGSEGV and just 
silently fails to use the PCH file.

> I was assuming that the cause of the problem was a mmap issue.  This 
> was based on the fact that the largefile.c test fails with 2.4 linux 
> kernels.  This is also a problem under hpux.  I haven't figured out 
> what to do with this.

I don't fully understand HPUX, but how could a `mmap issue' cause a 
function address to change?

Are you sure that largefile.c fails on linux?  It's not supposed to, 
rth wrote the required support. No-one has yet written the required 
support for HPUX, so it's not surprising that it fails.

> I'm fairly certain that I did a clean build.  If the PCH file depends
> on the compiler executable, this suggests that we aren't using the
> newly built compiler to build the PCH file, or we aren't using the
> PCH file that we think we are.

Sounds like something worth investigating.