perf_event: Fix incorrect range check on cpu number
authorPaul Mackerras <paulus@samba.org>
Tue, 15 Dec 2009 08:40:32 +0000 (19:40 +1100)
committerIngo Molnar <mingo@elte.hu>
Tue, 15 Dec 2009 12:09:55 +0000 (13:09 +0100)
commit0f624e7e5625f4c30c836b7a5decfe2553582391
tree6de6f9fbb15db70007238c923efb48dcde194c20
parenta6005123ce22770dbd91bc3cb637ce0807ab959b
perf_event: Fix incorrect range check on cpu number

It is quite legitimate for CPUs to be numbered sparsely, meaning
that it possible for an online CPU to have a number which is
greater than the total count of possible CPUs.

Currently find_get_context() has a sanity check on the cpu
number where it checks it against num_possible_cpus().  This
test can fail for a legitimate cpu number if the
cpu_possible_mask is sparsely populated.

This fixes the problem by checking the CPU number against
nr_cpumask_bits instead, since that is the appropriate check to
ensure that the cpu number is same to pass to cpu_isset()
subsequently.

Reported-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Tested-by: Michael Neuling <mikey@neuling.org>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: <stable@kernel.org>
LKML-Reference: <20091215084032.GA18661@brick.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/perf_event.c