Messages in this thread | | | Date | Tue, 13 Dec 2005 23:14:20 -0500 | From | Nathan Lynch <> | Subject | Re: [PATCH 1/2] Export cpu info by sysfs |
| |
Hello-
Zhang, Yanmin wrote: > I worked out 2 patches to export cpu topology and cache info by sysfs. > > The first patch is to export cpu topology info including below items > (attributes) which are similar to /proc/cpuinfo. > > /sys/devices/system/cpu/cpuX/topology/physical_package_id(representing > the physical package id of cpu X) > /sys/devices/system/cpu/cpuX/topology/core_id (representing the cpu core > id to cpu X) > /sys/devices/system/cpu/cpuX/topology/thread_id (representing the cpu > thread id to cpu X) > /sys/devices/system/cpu/cpuX/topology/thread_siblings (representing the > thread siblings to cpu X) > /sys/devices/system/cpu/cpuX/topology/core_siblings (represeting the > core siblings to cpu X)
I haven't looked at the patches in detail, but I have a concern about this approach. How is it that making new architecture-specific attributes under cpu directories in sysfs is preferable to the already architecture-specific format of /proc/cpuinfo and other proc entries?
If we're going to create a new user interface for exposing system topology (cores and threads etc), I would like for it to be as architecture-neutral as possible. We already do this for numa, for example.
Nathan - 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/
| |