On Thu, Jul 19, 2001 at 10:51:37PM +0200, Florian Lohoff wrote:
> On Thu, Jul 19, 2001 at 10:24:06PM +0200, Klaus Naumann wrote:
> > I know the problem but no solution. I suspect that it's a
> > problem of the poll function in Big Endian environments, because
> > I can reproduce this on my Indigo2 and on an Ultra 1 as well.
>
> I remember seeing a patch concerning this problem - Something with
> rtsignals - But i cant seem to find it anymore.
That's probably this one. I knew I'd let a patch slip. Glibc folk, is
this OK? The siginfo struct is different on MIPS.
--
Daniel Jacobowitz Carnegie Mellon University
MontaVista Software Debian GNU/Linux Developer
--- glibc-2.2.3/sysdeps/unix/sysv/linux/mips/bits/siginfo.h.orig Thu May 24 15:35:42 2001
+++ glibc-2.2.3/sysdeps/unix/sysv/linux/mips/bits/siginfo.h Thu May 24 15:35:49 2001
@@ -45,9 +45,9 @@
typedef struct siginfo
{
int si_signo; /* Signal number. */
+ int si_code; /* Signal code. */
int si_errno; /* If non-zero, an errno value associated with
this signal, as defined in <errno.h>. */
- int si_code; /* Signal code. */
union
{
