According commit d640113fe(ACPI: processor: fix acpi_get_cpuid for UP
processor), Bios may not provide _MAT or MADT tables and acpi_get_apicid()
always returns -1. For these cases, original code will pass apic_id with
vaule of -1 to acpi_map_cpuid() and it will check the acpi_id. If acpi_id
is equal to zero, ignores apic_id and return zero for CPU0.
Commit b981513(ACPI / scan: bail out early if failed to parse APIC
ID for CPU) changed the behavior. Return ENODEV when find apic_id is
less than zero after calling acpi_get_apicid(). This causes acpi-cpufreq
driver fails to be loaded on some machines. This patch is to fix it.
Reference:https://bugzilla.kernel.org/show_bug.cgi?id=73781
Cc: [email protected] v3.14
Reported-and-tested-by: KATO Hiroshi <[email protected]>
Reported-and-tested-by: Stuart Foster <[email protected]>
Signed-off-by: Lan Tianyu <[email protected]>
---
drivers/acpi/acpi_processor.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_processor.c
index c29c2c3..d55b603 100644
--- a/drivers/acpi/acpi_processor.c
+++ b/drivers/acpi/acpi_processor.c
@@ -260,10 +260,8 @@ static int acpi_processor_get_info(struct acpi_device *device)
}
apic_id = acpi_get_apicid(pr->handle, device_declaration, pr->acpi_id);
- if (apic_id < 0) {
+ if (apic_id < 0)
acpi_handle_debug(pr->handle, "failed to get CPU APIC ID.\n");
- return -ENODEV;
- }
pr->apic_id = apic_id;
cpu_index = acpi_map_cpuid(pr->apic_id, pr->acpi_id);
--
1.8.4.rc0.1.g8f6a3e5.dirty
Hi Tianyu,
With the suggested fix, we still need to protect
acpi_processor_hotadd_init() when pr->apic_id is -1,
otherwise it may cause invalid memory access.
Best Regards!
Gerry
diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_processor.c
index c29c2c3ec0ad..e17befc54c34 100644
--- a/drivers/acpi/acpi_processor.c
+++ b/drivers/acpi/acpi_processor.c
@@ -170,6 +170,9 @@ static int acpi_processor_hotadd_init(struct
acpi_processor *pr)
acpi_status status;
int ret;
+ if (pr->apic_id < 0)
+ return -ENODEV;
+
status = acpi_evaluate_integer(pr->handle, "_STA", NULL, &sta);
if (ACPI_FAILURE(status) || !(sta & ACPI_STA_DEVICE_PRESENT))
return -ENODEV;
On 2014/4/30 11:10, Lan Tianyu wrote:
> According commit d640113fe(ACPI: processor: fix acpi_get_cpuid for UP
> processor), Bios may not provide _MAT or MADT tables and acpi_get_apicid()
> always returns -1. For these cases, original code will pass apic_id with
> vaule of -1 to acpi_map_cpuid() and it will check the acpi_id. If acpi_id
> is equal to zero, ignores apic_id and return zero for CPU0.
>
> Commit b981513(ACPI / scan: bail out early if failed to parse APIC
> ID for CPU) changed the behavior. Return ENODEV when find apic_id is
> less than zero after calling acpi_get_apicid(). This causes acpi-cpufreq
> driver fails to be loaded on some machines. This patch is to fix it.
>
> Reference:https://bugzilla.kernel.org/show_bug.cgi?id=73781
> Cc: [email protected] v3.14
> Reported-and-tested-by: KATO Hiroshi <[email protected]>
> Reported-and-tested-by: Stuart Foster <[email protected]>
> Signed-off-by: Lan Tianyu <[email protected]>
> ---
> drivers/acpi/acpi_processor.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_processor.c
> index c29c2c3..d55b603 100644
> --- a/drivers/acpi/acpi_processor.c
> +++ b/drivers/acpi/acpi_processor.c
> @@ -260,10 +260,8 @@ static int acpi_processor_get_info(struct acpi_device *device)
> }
>
> apic_id = acpi_get_apicid(pr->handle, device_declaration, pr->acpi_id);
> - if (apic_id < 0) {
> + if (apic_id < 0)
> acpi_handle_debug(pr->handle, "failed to get CPU APIC ID.\n");
> - return -ENODEV;
> - }
> pr->apic_id = apic_id;
>
> cpu_index = acpi_map_cpuid(pr->apic_id, pr->acpi_id);
>
Hi Tianyu,
Found another issue at the second glance.
In case of x2apic, "if (apic_id < 0)" should be replaced
by "if (apic_id == -1)" for safety.
Best Regards!
Gerry
On 2014/4/30 13:08, Jiang Liu wrote:
> Hi Tianyu,
> With the suggested fix, we still need to protect
> acpi_processor_hotadd_init() when pr->apic_id is -1,
> otherwise it may cause invalid memory access.
>
> Best Regards!
> Gerry
>
> diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_processor.c
> index c29c2c3ec0ad..e17befc54c34 100644
> --- a/drivers/acpi/acpi_processor.c
> +++ b/drivers/acpi/acpi_processor.c
> @@ -170,6 +170,9 @@ static int acpi_processor_hotadd_init(struct
> acpi_processor *pr)
> acpi_status status;
> int ret;
>
> + if (pr->apic_id < 0)
> + return -ENODEV;
> +
> status = acpi_evaluate_integer(pr->handle, "_STA", NULL, &sta);
> if (ACPI_FAILURE(status) || !(sta & ACPI_STA_DEVICE_PRESENT))
> return -ENODEV;
>
>
> On 2014/4/30 11:10, Lan Tianyu wrote:
>> According commit d640113fe(ACPI: processor: fix acpi_get_cpuid for UP
>> processor), Bios may not provide _MAT or MADT tables and acpi_get_apicid()
>> always returns -1. For these cases, original code will pass apic_id with
>> vaule of -1 to acpi_map_cpuid() and it will check the acpi_id. If acpi_id
>> is equal to zero, ignores apic_id and return zero for CPU0.
>>
>> Commit b981513(ACPI / scan: bail out early if failed to parse APIC
>> ID for CPU) changed the behavior. Return ENODEV when find apic_id is
>> less than zero after calling acpi_get_apicid(). This causes acpi-cpufreq
>> driver fails to be loaded on some machines. This patch is to fix it.
>>
>> Reference:https://bugzilla.kernel.org/show_bug.cgi?id=73781
>> Cc: [email protected] v3.14
>> Reported-and-tested-by: KATO Hiroshi <[email protected]>
>> Reported-and-tested-by: Stuart Foster <[email protected]>
>> Signed-off-by: Lan Tianyu <[email protected]>
>> ---
>> drivers/acpi/acpi_processor.c | 4 +---
>> 1 file changed, 1 insertion(+), 3 deletions(-)
>>
>> diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_processor.c
>> index c29c2c3..d55b603 100644
>> --- a/drivers/acpi/acpi_processor.c
>> +++ b/drivers/acpi/acpi_processor.c
>> @@ -260,10 +260,8 @@ static int acpi_processor_get_info(struct acpi_device *device)
>> }
>>
>> apic_id = acpi_get_apicid(pr->handle, device_declaration, pr->acpi_id);
>> - if (apic_id < 0) {
>> + if (apic_id < 0)
>> acpi_handle_debug(pr->handle, "failed to get CPU APIC ID.\n");
>> - return -ENODEV;
>> - }
>> pr->apic_id = apic_id;
>>
>> cpu_index = acpi_map_cpuid(pr->apic_id, pr->acpi_id);
>>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
On 2014年04月30日 13:16, Jiang Liu wrote:
> Hi Tianyu,
> Found another issue at the second glance.
> In case of x2apic, "if (apic_id < 0)" should be replaced
> by "if (apic_id == -1)" for safety.
Hi Jiang:
Thanks for suggestion. I will update soon.
> Best Regards!
> Gerry
>
> On 2014/4/30 13:08, Jiang Liu wrote:
>> Hi Tianyu,
>> With the suggested fix, we still need to protect
>> acpi_processor_hotadd_init() when pr->apic_id is -1,
>> otherwise it may cause invalid memory access.
>>
>> Best Regards!
>> Gerry
>>
>> diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_processor.c
>> index c29c2c3ec0ad..e17befc54c34 100644
>> --- a/drivers/acpi/acpi_processor.c
>> +++ b/drivers/acpi/acpi_processor.c
>> @@ -170,6 +170,9 @@ static int acpi_processor_hotadd_init(struct
>> acpi_processor *pr)
>> acpi_status status;
>> int ret;
>>
>> + if (pr->apic_id < 0)
>> + return -ENODEV;
>> +
>> status = acpi_evaluate_integer(pr->handle, "_STA", NULL, &sta);
>> if (ACPI_FAILURE(status) || !(sta & ACPI_STA_DEVICE_PRESENT))
>> return -ENODEV;
>>
>>
>> On 2014/4/30 11:10, Lan Tianyu wrote:
>>> According commit d640113fe(ACPI: processor: fix acpi_get_cpuid for UP
>>> processor), Bios may not provide _MAT or MADT tables and acpi_get_apicid()
>>> always returns -1. For these cases, original code will pass apic_id with
>>> vaule of -1 to acpi_map_cpuid() and it will check the acpi_id. If acpi_id
>>> is equal to zero, ignores apic_id and return zero for CPU0.
>>>
>>> Commit b981513(ACPI / scan: bail out early if failed to parse APIC
>>> ID for CPU) changed the behavior. Return ENODEV when find apic_id is
>>> less than zero after calling acpi_get_apicid(). This causes acpi-cpufreq
>>> driver fails to be loaded on some machines. This patch is to fix it.
>>>
>>> Reference:https://bugzilla.kernel.org/show_bug.cgi?id=73781
>>> Cc: [email protected] v3.14
>>> Reported-and-tested-by: KATO Hiroshi <[email protected]>
>>> Reported-and-tested-by: Stuart Foster <[email protected]>
>>> Signed-off-by: Lan Tianyu <[email protected]>
>>> ---
>>> drivers/acpi/acpi_processor.c | 4 +---
>>> 1 file changed, 1 insertion(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_processor.c
>>> index c29c2c3..d55b603 100644
>>> --- a/drivers/acpi/acpi_processor.c
>>> +++ b/drivers/acpi/acpi_processor.c
>>> @@ -260,10 +260,8 @@ static int acpi_processor_get_info(struct acpi_device *device)
>>> }
>>>
>>> apic_id = acpi_get_apicid(pr->handle, device_declaration, pr->acpi_id);
>>> - if (apic_id < 0) {
>>> + if (apic_id < 0)
>>> acpi_handle_debug(pr->handle, "failed to get CPU APIC ID.\n");
>>> - return -ENODEV;
>>> - }
>>> pr->apic_id = apic_id;
>>>
>>> cpu_index = acpi_map_cpuid(pr->apic_id, pr->acpi_id);
>>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>> the body of a message to [email protected]
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>> Please read the FAQ at http://www.tux.org/lkml/
>>
--
Best regards
Tianyu Lan