Of all the reasons that dlpar_cpu_remove() can fail, the 'last online CPU' is one that can be caused directly by the user offlining CPUs in a partition/virtual machine that has hotplugged CPUs. Trying to reclaim a hotplugged CPU can fail if the CPU is now the last online in the system. This is easily reproduced using QEMU [1].
Throwing a more specific error message for this case, instead of just "Failed to offline CPU", makes it clearer that the error is in fact a known error situation instead of other generic/unknown cause. [1] https://bugzilla.redhat.com/1911414 Signed-off-by: Daniel Henrique Barboza <danielhb...@gmail.com> --- arch/powerpc/platforms/pseries/hotplug-cpu.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c index 12cbffd3c2e3..134f393f09e1 100644 --- a/arch/powerpc/platforms/pseries/hotplug-cpu.c +++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c @@ -514,7 +514,17 @@ static ssize_t dlpar_cpu_remove(struct device_node *dn, u32 drc_index) rc = dlpar_offline_cpu(dn); if (rc) { - pr_warn("Failed to offline CPU %pOFn, rc: %d\n", dn, rc); + /* dlpar_offline_cpu will return -EBUSY from cpu_down() (via + * device_offline()) in 2 cases: cpu_hotplug_disable is true or + * there is only one CPU left. Warn the user about the second + * since this can happen with user offlining CPUs and then + * attempting hotunplugs. + */ + if (rc == -EBUSY && num_online_cpus() == 1) + pr_warn("Unable to remove last online CPU %pOFn\n", dn); + else + pr_warn("Failed to offline CPU %pOFn, rc: %d\n", dn, rc); + return -EINVAL; } -- 2.29.2
