BUG #4448: postgres won't start with latest stable kernel

The following bug has been logged online:

Bug reference:      4448
Logged by:          Donald Harter
Email address:      harterc1@comcast.net
PostgreSQL version: 8.3.1-15.1
Operating system:   SUSE11.0
Description:        postgres won't start with latest stable kernel
Details:

When I use this kernel postgres won't start.
Linux linux-dzdg 2.6.26.5-default #1 SMP Tue Sep 23 05:12:39 CDT 2008 x86_64
x86_64 x86_64 GNU/Linux
When I fall back to an earlier kernel version such as linux-2.6.25.16-0.1 I
can start postgres.  In both cases I run the same command. "postgres -d 5 -D
/srv/db -i -h localhost -r logout"

I suspect this is a kernel issue.  When I look at the latest unstable kernel
changes I see some references to system calls which fail when I try to
starrt postgres.  Using strace I have:
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 3
setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
bind(3, {sa_family=AF_INET, sin_port=htons(5432),
sin_addr=inet_addr("127.0.0.1")}, 16) = 0
listen(3, 206)                          = 0
open("/tmp/.s.PGSQL.5432.lock", O_RDWR|O_CREAT|O_EXCL, 0600) = 4
write(4, "2051\n/srv/db\n", 13)         = 13
close(4)                                = 0
unlink("/tmp/.s.PGSQL.5432")            = -1 ENOENT (No such file or
directory)
socket(PF_FILE, SOCK_STREAM, 0)         = 4
bind(4, {sa_family=AF_FILE, path="/tmp/.s.PGSQL.5432"}, 110) = 0
chmod("/tmp/.s.PGSQL.5432", 0777)       = 0
listen(4, 206)                          = 0
write(2, "2008-10-03 10:20:31 CDT   DEBUG:"..., 74) = 74
shmget(0x52e2c1, 39288832, IPC_CREAT|IPC_EXCL|0600) = -1 EINVAL (Invalid
argument)
write(2, "2008-10-03 10:20:31 CDT   FATAL:"..., 878) = 878
write(2, "2008-10-03 10:20:31 CDT   DEBUG:"..., 47) = 47
write(2, "2008-10-03 10:20:31 CDT   DEBUG:"..., 48) = 48
unlink("/tmp/.s.PGSQL.5432")            = 0
unlink("/tmp/.s.PGSQL.5432.lock")       = 0
unlink("postmaster.pid")                = 0
write(2, "2008-10-03 10:20:31 CDT   DEBUG:"..., 42) = 42
exit_group(1)

The call to shmget fails and I suspect that is the problem.