projects / linux-2.6-block.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 394ffa5) | patch
blk-mq: use 'nr_cpu_ids' as highest CPU ID count for hwq <-> cpu map
authorJens Axboe <axboe@fb.com>
Mon, 24 Nov 2014 22:02:42 +0000 (15:02 -0700)
committerJens Axboe <axboe@fb.com>
Mon, 24 Nov 2014 22:02:42 +0000 (15:02 -0700)
commita33c1ba2913802b6fb23e974bb2f6a4e73c8b7ce
treef1dbb27e9804ed4f8368e4eb0169b7325ec0792ftree | snapshot
parent394ffa503bc40e32d7f54a9b817264e81ce131b4commit | diff
blk-mq: use 'nr_cpu_ids' as highest CPU ID count for hwq <-> cpu map

We currently use num_possible_cpus(), but that breaks on sparc64 where
the CPU ID space is discontig. Use nr_cpu_ids as the highest CPU ID
instead, so we don't end up reading from invalid memory.

Cc: stable@kernel.org # 3.13+
Signed-off-by: Jens Axboe <axboe@fb.com>
block/blk-mq-cpumap.c diff | blob | blame | history
Linux 6.x block layer and io_uring tree(s)