Revision be14eb619108fa8b7120eb2c42d66d5f623ae10e authored by Brian King on 10 September 2010, 07:03:21 UTC, committed by Jens Axboe on 10 September 2010, 07:03:21 UTC
While testing CPU DLPAR, the following problem was discovered.
We were DLPAR removing the first CPU, which in this case was
logical CPUs 0-3. CPUs 0-2 were already marked offline and
we were in the process of offlining CPU 3. After marking
the CPU inactive and offline in cpu_disable, but before the
cpu was completely idle (cpu_die), we ended up in __make_request
on CPU 3. There we looked at the topology map to see which CPU
to complete the I/O on and found no CPUs in the cpu_sibling_map.
This resulted in the block layer setting the completion cpu
to be NR_CPUS, which then caused an oops when we tried to
complete the I/O.

Fix this by sanity checking the value we return from blk_cpu_to_group
to be a valid cpu value.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
1 parent edce682
History
File Mode Size
Kconfig -rw-r--r-- 6.4 KB
Makefile -rw-r--r-- 1.0 KB
ip6_queue.c -rw-r--r-- 14.0 KB
ip6_tables.c -rw-r--r-- 57.9 KB
ip6t_LOG.c -rw-r--r-- 12.1 KB
ip6t_REJECT.c -rw-r--r-- 6.6 KB
ip6t_ah.c -rw-r--r-- 3.1 KB
ip6t_eui64.c -rw-r--r-- 1.8 KB
ip6t_frag.c -rw-r--r-- 3.9 KB
ip6t_hbh.c -rw-r--r-- 5.0 KB
ip6t_ipv6header.c -rw-r--r-- 3.5 KB
ip6t_mh.c -rw-r--r-- 2.3 KB
ip6t_rt.c -rw-r--r-- 5.8 KB
ip6table_filter.c -rw-r--r-- 3.0 KB
ip6table_mangle.c -rw-r--r-- 4.0 KB
ip6table_raw.c -rw-r--r-- 2.1 KB
ip6table_security.c -rw-r--r-- 2.8 KB
nf_conntrack_l3proto_ipv6.c -rw-r--r-- 12.6 KB
nf_conntrack_proto_icmpv6.c -rw-r--r-- 8.9 KB
nf_conntrack_reasm.c -rw-r--r-- 16.3 KB

back to top