Messages in this thread |  | | | Subject | Re: the creation of boot_cpu_init() is wrong and accessing uninitialised data | | From | James Bottomley <> | | Date | Tue, 27 Jun 2006 09:01:15 -0500 |
| |
On Tue, 2006-06-27 at 23:00 +1000, Keith Owens wrote: > AFAICR, the BSP is supposed to be logical cpu 0 on all architectures. > Most architectures assign logical cpu 0 to the BSP, even if the BSP > has > a non-zero hard_smp_processor_id. ia64 even has this
That's definitely not a requirement. For systems like voyager whose CPUs cannot be renumbered it makes no sense. The original reason for renumbering was that most CPU loops ran from 1 to max_cpu and therefore worked better if the CPU map was dense. However, I fixed that up long ago so sparse CPU map traversal should be fine now.
James
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|  |