powerpc, hotplug: Avoid to touch non-existent cpumasks.

We observed a kernel oops when running a PPC guest with config NR_CPUS=4
and qemu option "-smp cores=1,threads=8":

[   30.634781] Unable to handle kernel paging request for data at
address 0xc00000014192eb17
[   30.636173] Faulting instruction address: 0xc00000000003e5cc
[   30.637069] Oops: Kernel access of bad area, sig: 11 [#1]
[   30.637877] SMP NR_CPUS=4 NUMA pSeries
[   30.638471] Modules linked in:
[   30.638949] CPU: 3 PID: 27 Comm: migration/3 Not tainted
4.7.0-07963-g9714b26 #1
[   30.640059] task: c00000001e29c600 task.stack: c00000001e2a8000
[   30.640956] NIP: c00000000003e5cc LR: c00000000003e550 CTR:
0000000000000000
[   30.642001] REGS: c00000001e2ab8e0 TRAP: 0300   Not tainted
(4.7.0-07963-g9714b26)
[   30.643139] MSR: 8000000102803033 <SF,VEC,VSX,FP,ME,IR,DR,RI,LE,TM[E]>  CR: 22004084  XER: 00000000
[   30.644583] CFAR: c000000000009e98 DAR: c00000014192eb17 DSISR: 40000000 SOFTE: 0
GPR00: c00000000140a6b8 c00000001e2abb60 c0000000016dd300 0000000000000003
GPR04: 0000000000000000 0000000000000004 c0000000016e5920 0000000000000008
GPR08: 0000000000000004 c00000014192eb17 0000000000000000 0000000000000020
GPR12: c00000000140a6c0 c00000000ffffc00 c0000000000d3ea8 c00000001e005680
GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
GPR20: 0000000000000000 c00000001e6b3a00 0000000000000000 0000000000000001
GPR24: c00000001ff85138 c00000001ff85130 000000001eb6f000 0000000000000001
GPR28: 0000000000000000 c0000000017014e0 0000000000000000 0000000000000018
[   30.653882] NIP [c00000000003e5cc] __cpu_disable+0xcc/0x190
[   30.654713] LR [c00000000003e550] __cpu_disable+0x50/0x190
[   30.655528] Call Trace:
[   30.655893] [c00000001e2abb60] [c00000000003e550] __cpu_disable+0x50/0x190 (unreliable)
[   30.657280] [c00000001e2abbb0] [c0000000000aca0c] take_cpu_down+0x5c/0x100
[   30.658365] [c00000001e2abc10] [c000000000163918] multi_cpu_stop+0x1a8/0x1e0
[   30.659617] [c00000001e2abc60] [c000000000163cc0] cpu_stopper_thread+0xf0/0x1d0
[   30.660737] [c00000001e2abd20] [c0000000000d8d70] smpboot_thread_fn+0x290/0x2a0
[   30.661879] [c00000001e2abd80] [c0000000000d3fa8] kthread+0x108/0x130
[   30.662876] [c00000001e2abe30] [c000000000009968] ret_from_kernel_thread+0x5c/0x74
[   30.664017] Instruction dump:
[   30.664477] 7bde1f24 38a00000 787f1f24 3b600001 39890008 7d204b78 7d05e214 7d0b07b4
[   30.665642] 796b1f24 7d26582a 7d204a14 7d29f214 <7d4048a8> 7d4a3878 7d4049ad 40c2fff4
[   30.666854] ---[ end trace 32643b7195717741 ]---

The reason of this is that in __cpu_disable(), when we try to set the
cpu_sibling_mask or cpu_core_mask of the sibling CPUs of the disabled
one, we don't check whether the current configuration employs those
sibling CPUs(hw threads). And if a CPU is not employed by a
configuration, the percpu structures cpu_{sibling,core}_mask are not
allocated, therefore accessing those cpumasks will result in problems as
above.

This patch fixes this problem by adding an addition check on whether the
id is no less than nr_cpu_ids in the sibling CPU iteration code.

Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 25a39052bf6b5653c3ab0317298864a2873fd911..9c6f3fd580597e5fdfcc5fc46031d9be0aaac74b 100644 (file)
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -830,7 +830,7 @@ int __cpu_disable(void)
 
        /* Update sibling maps */
        base = cpu_first_thread_sibling(cpu);
-       for (i = 0; i < threads_per_core; i++) {
+       for (i = 0; i < threads_per_core && base + i < nr_cpu_ids; i++) {
                cpumask_clear_cpu(cpu, cpu_sibling_mask(base + i));
                cpumask_clear_cpu(base + i, cpu_sibling_mask(cpu));
                cpumask_clear_cpu(cpu, cpu_core_mask(base + i));