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-- 2.5 KB
Kconfig.iosched -rw-r--r-- 1.8 KB
Makefile -rw-r--r-- 592 bytes
blk-barrier.c -rw-r--r-- 8.6 KB
blk-cgroup.c -rw-r--r-- 26.8 KB
blk-cgroup.h -rw-r--r-- 8.0 KB
blk-core.c -rw-r--r-- 69.0 KB
blk-exec.c -rw-r--r-- 2.6 KB
blk-integrity.c -rw-r--r-- 10.0 KB
blk-ioc.c -rw-r--r-- 3.9 KB
blk-iopoll.c -rw-r--r-- 5.9 KB
blk-lib.c -rw-r--r-- 5.1 KB
blk-map.c -rw-r--r-- 8.1 KB
blk-merge.c -rw-r--r-- 10.6 KB
blk-settings.c -rw-r--r-- 25.1 KB
blk-softirq.c -rw-r--r-- 4.1 KB
blk-sysfs.c -rw-r--r-- 14.0 KB
blk-tag.c -rw-r--r-- 9.9 KB
blk-timeout.c -rw-r--r-- 5.6 KB
blk.h -rw-r--r-- 4.6 KB
bsg.c -rw-r--r-- 23.5 KB
cfq-iosched.c -rw-r--r-- 103.7 KB
cfq.h -rw-r--r-- 3.6 KB
compat_ioctl.c -rw-r--r-- 21.0 KB
deadline-iosched.c -rw-r--r-- 11.4 KB
elevator.c -rw-r--r-- 25.4 KB
genhd.c -rw-r--r-- 29.9 KB
ioctl.c -rw-r--r-- 8.0 KB
noop-iosched.c -rw-r--r-- 2.6 KB
scsi_ioctl.c -rw-r--r-- 18.0 KB

back to top