Greg KH wrote:
On Thu, Mar 19, 2009 at 10:36:38AM +0900, Kenji Kaneshige wrote:
Greg KH wrote:
On Wed, Mar 18, 2009 at 05:29:11PM +0900, Taku Izumi wrote:
Hi Greg, Alex
This patchset adds an uevent framework to pci_hotplug.
Why?
What is this needed for?
Who is going to use this?
Now there is no hotplug management utility as far as I know.
Not true, I know of at least 2 for PCI, one opensource and one not.
I think such utility is necessary for system administrators and
the hotplug event notification is useful for it.
Nope, it's not necessary.
Do you mean utility doesn't need to handle hotplug events, or
utility can handle events without notification mechanism?
From what I remember, it was handled without a notification mechanism,
but that was years ago.
Ok, thank you. I'll look at the source code of pcihpview. I guess
it checks the status such as power, latch and adapter periodically.
With notification mechanism, it will not need periodical checks. In
addition, there are some events that we cannot check from the files
under slot directory, such as attention button event. Currently,
attention button events of shpchp or pciehp are notified to userland
only through kernel message. I think it's very poor.
Thanks,
Kenji Kaneshige
BTW, IIRC, PCI hotplug core used to have a event notification
mechanism using sysfs_update_file(*) and some people (maybe
including you) tried to utilize it in the userland ulitity (
pcihputil?). But it seems to have been given up. If you know
about that, could you tell me what the problem was?
Sorry, I really don't remember.
thanks,
greg k-h
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
