RE: [PATCH v3]Export cpu topology by sysfs

From: Zhang, Yanmin <>
Date: 2006-01-09 17:47:48
>>-----Original Message-----
>>From: [] On Behalf Of Nathan Lynch
>>Sent: 200619 13:10
>>To: Yanmin Zhang
>>Cc:;;;; Siddha, Suresh B; Shah, Rajesh;
>>Pallipadi, Venkatesh
>>Subject: Re: [PATCH v3]Export cpu topology by sysfs
>>> 2) Set consistent default values for the 4 attributes.
>>> If one architecture wants to support this feature, it just needs to
>>> implement 4 defines, typically in file include/asm-XXX/topology.h.
>>> The 4 defines are:
>>> #define topology_physical_package_id(cpu)
>>> #define topology_core_id(cpu)
>>> #define topology_thread_siblings(cpu)
>>> #define topology_core_siblings(cpu)
>>> The type of **_id is int.
>>> The type of siblings is cpumask_t.
>>> To be consistent on all architectures, the 4 attributes should have
>>> deafult values if their values are unavailable. Below is the rule.
>>> 1) physical_package_id: If cpu has no physical package id, -1 is the
>>> default value.
>>> 2) core_id: If cpu doesn't support multi-core, its core id is 0.
>>Why not -1 as with the physical package id?  0 could be a valid core
If the cpu has only 1 core, we could call it single-core, so it's reasonable to use 0 as its core id.

>>> 3) thread_siblings: Just include itself, if the cpu doesn't support
>>> HT/multi-thread.
>>> 4) core_siblings: Just include itself, if the cpu doesn't support
>>> multi-core and HT/Multi-thread.
>>Really, I think the least confusing interface would not export those
>>attributes which are not relevant for the system.  E.g. if the system
>>isn't multi-core, you don't see core_id and core_siblings attributes.
>>Failing that, let's at least have consistent, unambiguous values for
>>the ids which are not applicable.
Current kernel will output core id by /proc/cpuinfo if a physical cpu has 2 threads, no matter if it's a multi-core, or just a multi-thread. To be consistent with /proc/cpuinfo, I think we need export core id and its default value is 0.

>>> +static int __cpuinit topology_cpu_callback(struct notifier_block *nfb,
>>> +		unsigned long action, void *hcpu)
>>> +{
>>> +	unsigned int cpu = (unsigned long)hcpu;
>>> +	struct sys_device *sys_dev;
>>> +
>>> +	sys_dev = get_cpu_sysdev(cpu);
>>> +	switch (action) {
>>> +		case CPU_ONLINE:
>>> +			topology_add_dev(sys_dev);
>>> +			break;
>>> +		case CPU_DEAD:
>>> +			topology_remove_dev(sys_dev);
>>> +			break;
>>> +	}
>>> +	return NOTIFY_OK;
>>> +}
>>I still oppose this bit.  I want the attributes there for powerpc even
>>for offline cpus -- the topology information (if obtainable, which it
>>is on powerpc) is useful regardless of the cpu's online state.  The
>>attributes should appear when a cpu is made present, and go away when
>>a cpu is removed.
As my previous email says, there are concerns/issues to do so. A compromise is that the patch could register a sysdev driver. When the cpu becomes offline from online, we don't delete the topology kobj. The compromise has a defect. If the cpu is never online since machine boots, the topology info of the cpu is incorrect. 

>>This week I'll try to do an implementation for powerpc.
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to
More majordomo info at
Received on Mon Jan 09 17:48:35 2006

This archive was generated by hypermail 2.1.8 : 2006-01-09 17:48:43 EST