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
File | Mode | Size |
---|---|---|
Kconfig | -rw-r--r-- | 898 bytes |
Makefile | -rw-r--r-- | 416 bytes |
afs.h | -rw-r--r-- | 6.0 KB |
afs_cm.h | -rw-r--r-- | 1.2 KB |
afs_fs.h | -rw-r--r-- | 2.3 KB |
afs_vl.h | -rw-r--r-- | 3.6 KB |
cache.c | -rw-r--r-- | 10.8 KB |
callback.c | -rw-r--r-- | 11.4 KB |
cell.c | -rw-r--r-- | 10.3 KB |
cmservice.c | -rw-r--r-- | 13.2 KB |
dir.c | -rw-r--r-- | 27.1 KB |
file.c | -rw-r--r-- | 8.6 KB |
flock.c | -rw-r--r-- | 15.8 KB |
fsclient.c | -rw-r--r-- | 44.1 KB |
inode.c | -rw-r--r-- | 12.1 KB |
internal.h | -rw-r--r-- | 30.9 KB |
main.c | -rw-r--r-- | 4.2 KB |
misc.c | -rw-r--r-- | 2.3 KB |
mntpt.c | -rw-r--r-- | 7.5 KB |
netdevices.c | -rw-r--r-- | 1.5 KB |
proc.c | -rw-r--r-- | 17.0 KB |
rxrpc.c | -rw-r--r-- | 20.1 KB |
security.c | -rw-r--r-- | 8.9 KB |
server.c | -rw-r--r-- | 7.9 KB |
super.c | -rw-r--r-- | 12.3 KB |
vlclient.c | -rw-r--r-- | 5.4 KB |
vlocation.c | -rw-r--r-- | 17.7 KB |
vnode.c | -rw-r--r-- | 24.1 KB |
volume.c | -rw-r--r-- | 9.9 KB |
write.c | -rw-r--r-- | 18.4 KB |
Computing file changes ...