This is the mail archive of the
mailing list for the glibc project.
Re: changiing rtld.c
- From: Aldy Hernandez <aldyh at redhat dot com>
- To: Steve Munroe <sjmunroe at us dot ibm dot com>
- Cc: libc-alpha at sources dot redhat dot com, Peter Bergner <bergner at us dot ibm dot com>
- Date: Tue, 1 Oct 2002 15:53:31 -0700
- Subject: Re: changiing rtld.c
- References: <OFCB948751.43689757-ON86256C45.007D5268@rchland.ibm.com>
> The ppc32 kernel (and current ppc64 kernels) does not handle SIGFPE
> correctly which causes a loop if the signal handler touches the FP unit.
> You will need a kernel patch. Peter Bergner is working on the fix. I will
> talk to him tomorrow.
:-) Lemme give a little more background.
I'm working on patches to glibc so powerpc can be configured entirely
with soft-float (exceptions and everything simulated). They use the
glibc/soft-float/* functionality, plus some glue code I've written.
So... The SIGFPE is definitely being caused by my hacks, not by the
kernel or the hardware :).
I just want to track down where exactly it got generated.