5.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jonathan Cameron Jonathan.Cameron@huawei.com
[ Upstream commit 2488444274c70038eb6b686cba5f1ce48ebb9cdd ]
In a review discussion of the changes to support vCPU hotplug where a check was added on the GICC being enabled if was online, it was noted that there is need to map back to the cpu and use that to index into a cpumask. As such, a valid ID is needed.
If an MPIDR check fails in acpi_map_gic_cpu_interface() it is possible for the entry in cpu_madt_gicc[cpu] == NULL. This function would then cause a NULL pointer dereference. Whilst a path to trigger this has not been established, harden this caller against the possibility.
Reviewed-by: Gavin Shan gshan@redhat.com Signed-off-by: Jonathan Cameron Jonathan.Cameron@huawei.com Link: https://lore.kernel.org/r/20240529133446.28446-13-Jonathan.Cameron@huawei.co... Signed-off-by: Catalin Marinas catalin.marinas@arm.com Signed-off-by: Sasha Levin sashal@kernel.org --- arch/arm64/include/asm/acpi.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/include/asm/acpi.h b/arch/arm64/include/asm/acpi.h index 0d1da93a5bad..702587fda70c 100644 --- a/arch/arm64/include/asm/acpi.h +++ b/arch/arm64/include/asm/acpi.h @@ -102,7 +102,8 @@ static inline int get_cpu_for_acpi_id(u32 uid) int cpu;
for (cpu = 0; cpu < nr_cpu_ids; cpu++) - if (uid == get_acpi_id_for_cpu(cpu)) + if (acpi_cpu_get_madt_gicc(cpu) && + uid == get_acpi_id_for_cpu(cpu)) return cpu;
return -EINVAL;