Jesse Barnes wrote:
On Tue, 28 Jul 2009 11:38:27 -0500
Mike Habeck <habeck@xxxxxxx> wrote:
Jesse Barnes wrote:
On Mon, 27 Jul 2009 12:04:21 +0900
Kenji Kaneshige <kaneshige.kenji@xxxxxxxxxxxxxx> wrote:
Hi,
I noticed the following regression problems in SGI hotplug driver.
I'm sending the patches for these problems. Unfortunately, I don't
have any hardware environment to test the patches, so I hope
someone test the patches on a appropreate environment.
- The commit bd3d99c17039fd05a29587db3f4a180c48da115a ("PCI:
Remove untested Electromechanical Interlock (EMI) support in
pciehp."), which removes the definition of "struct
hotplug_slot_attr", broke SGI hotplug driver because SGI hotplug
driver is still using struct hotplug_slot_attr.
- Adding the struct hotplug_slot_attr definition again is not
enough because hotplug_slot_attr no longer work. Maybe by the
recent pci slot changes, callbacks of attributes
under /sys/bus/pci/slots had been changed to get the pointer to
struct pci_slot instead of struct hotplug_slot. So the path_show()
that assumes the parameter is a pointer to struct hotplug_slot
seems broken.
The patches are:
- [PATCH 1/2] SGI hotplug: fix build failure
- [PATCH 2/2] SGI hotplug: do not use hotplug_slot_attr
Any issue with these Parit or Jes? I'll queue them up until I hear
someone yell.
Jesse,
I was going to build a kernel with the patch applied and test
it on an Altix later today...
Great, thanks Mike. Let me know if additional patches are needed or if
you want to do something different.
Thanks,
Jesse (Kenji),
I installed the patch and tested on a Altix 4700. It looks good.
Actually the resulting code of the two patches is the exact same
outcome as a patch that sgi had come up with and had in a local
tree, but somehow never got sent out to the community.
Kenji, Thanks for pushing this forward!
-mike
--
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
