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 |
---|---|---|
Makefile | -rw-r--r-- | 254 bytes |
af_rose.c | -rw-r--r-- | 38.3 KB |
rose_dev.c | -rw-r--r-- | 3.6 KB |
rose_in.c | -rw-r--r-- | 7.5 KB |
rose_link.c | -rw-r--r-- | 6.9 KB |
rose_loopback.c | -rw-r--r-- | 2.4 KB |
rose_out.c | -rw-r--r-- | 2.8 KB |
rose_route.c | -rw-r--r-- | 31.0 KB |
rose_subr.c | -rw-r--r-- | 11.6 KB |
rose_timer.c | -rw-r--r-- | 4.9 KB |
sysctl_net_rose.c | -rw-r--r-- | 3.3 KB |
Computing file changes ...