On Thu, Nov 23, 2023 at 03:36:17PM +0530, Raag Jadav wrote:
> Now that we have _UID matching support for integer types, we can use
> acpi_dev_hid_uid_match() for it.
>
> Signed-off-by: Raag Jadav <raag.jadav@xxxxxxxxx>
> Acked-by: Will Deacon <will@xxxxxxxxxx>
> ---
> drivers/perf/arm_cspmu/arm_cspmu.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/drivers/perf/arm_cspmu/arm_cspmu.c b/drivers/perf/arm_cspmu/arm_cspmu.c
> index 2cc35dded007..50b89b989ce7 100644
> --- a/drivers/perf/arm_cspmu/arm_cspmu.c
> +++ b/drivers/perf/arm_cspmu/arm_cspmu.c
> @@ -1108,7 +1108,6 @@ static int arm_cspmu_request_irq(struct arm_cspmu *cspmu)
>
> static inline int arm_cspmu_find_cpu_container(int cpu, u32 container_uid)
> {
> - u64 acpi_uid;
> struct device *cpu_dev;
> struct acpi_device *acpi_dev;
>
> @@ -1118,8 +1117,7 @@ static inline int arm_cspmu_find_cpu_container(int cpu, u32 container_uid)
>
> acpi_dev = ACPI_COMPANION(cpu_dev);
> while (acpi_dev) {
> - if (acpi_dev_hid_uid_match(acpi_dev, ACPI_PROCESSOR_CONTAINER_HID, NULL) &&
> - !acpi_dev_uid_to_integer(acpi_dev, &acpi_uid) && acpi_uid == container_uid)
> + if (acpi_dev_hid_uid_match(acpi_dev, ACPI_PROCESSOR_CONTAINER_HID, container_uid))
> return 0;
On second thought, I just realized that this can cause false positive
in case container_uid is integer 0.
I'm not aware if that's a valid value to match against in APMT.
If it is, it's better to drop patches 2 and 5 for now.
Raag
