On Tue, May 05, 2015 at 02:14:01PM +0100, Hanjun Guo wrote:
On 2015???05???05??? 19:25, Sudeep Holla wrote:
On 05/05/15 03:46, Hanjun Guo wrote:
Introduce invalid_phys_cpuid() to identify cpu with invalid physical ID, then used it as replacement of the direct comparisons with PHYS_CPUID_INVALID.
Signed-off-by: Hanjun Guo hanjun.guo@linaro.org
drivers/acpi/acpi_processor.c | 4 ++-- drivers/acpi/processor_core.c | 4 ++-- include/linux/acpi.h | 5 +++++ 3 files changed, 9 insertions(+), 4 deletions(-)
diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_processor.c index 62c846b..92a5f73 100644 --- a/drivers/acpi/acpi_processor.c
[...]
diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 913b49f..cc82ff3 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -163,6 +163,11 @@ static inline bool invalid_logical_cpuid(u32 cpuid) return (int)cpuid < 0; }
+static inline bool invalid_phys_cpuid(phys_cpuid_t phys_id) +{
- return (int)phys_id < 0;
Should this be phys_id == PHYS_CPUID_INVALID ? else I don't see why we need to even define PHYS_CPUID_INVALID
I'm OK with this. For now, CPU phys_id will be valid value or PHYS_CPUID_INVALID in all cases for ACPI processor driver, but I want ask Rafael's opinion on this, is it OK to you too, Rafael?
Is your worry related to functions returning error values other than PHYS_CPUID_INVALID (when they are expected to return a physical id) ? Is there any in the current kernel ?
static inline bool invalid_phys_cpuid(phys_cpuid_t phys_id) { return phys_id == PHYS_CPUID_INVALID; }
This should do, and if we need more mapping functions that are supposed to return physical ids they should return PHYS_CPUID_INVALID on failure.
Lorenzo