PNP: mark resources that conflict with PCI devices "disabled"

Both the PNP/PCI conflict detection quirk and the PNP system
driver must use the same mechanism to mark resources as disabled.

I think it's best to keep the resource and to keep the type bit
(IORESOURCE_MEM, etc), so that we match the list from firmware
as closely as possible.

Fixes this regression from 2.6.25: http://lkml.org/lkml/2008/6/1/82

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Tested-by: Avuton Olrich <avuton@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

2 files changed, 2 insertions, 2 deletions

diff --git a/drivers/pnp/quirks.c b/drivers/pnp/quirks.c
index e2b7de4cb05e..1ff3bb585ab2 100644
--- a/drivers/pnp/quirks.c
+++ b/drivers/pnp/quirks.c
@@ -286,7 +286,7 @@ static void quirk_system_pci_resources(struct pnp_dev *dev)
                                         pci_name(pdev), i,
                                         (unsigned long long) pci_start,
                                         (unsigned long long) pci_end);
-                                res->flags = 0;
+                                res->flags |= IORESOURCE_DISABLED;
                         }
                 }
         }
diff --git a/drivers/pnp/system.c b/drivers/pnp/system.c
index 9c2496dbeee4..8f0a570509c5 100644
--- a/drivers/pnp/system.c
+++ b/drivers/pnp/system.c
@@ -81,7 +81,7 @@ static void reserve_resources_of_dev(struct pnp_dev *dev)
         }
 
         for (i = 0; (res = pnp_get_resource(dev, IORESOURCE_MEM, i)); i++) {
-                if (res->flags & IORESOURCE_UNSET)
+                if (res->flags & IORESOURCE_DISABLED)
                         continue;
 
                 reserve_range(dev, res->start, res->end, 0);