Patchwork powerpc/pci: clean up direct access to sysdata by iseries platform

login
register
mail settings
Submitter Kumar Gala
Date May 5, 2009, 12:04 p.m.
Message ID <1241525095-18041-1-git-send-email-galak@kernel.crashing.org>
Download mbox | patch
Permalink /patch/26874/
State Superseded, archived
Headers show

Comments

Kumar Gala - May 5, 2009, 12:04 p.m.
We shouldn't directly access sysdata to get the device node.  We should
be calling pci_device_to_OF_node().

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/platforms/iseries/pci.c |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)
Stephen Rothwell - May 6, 2009, 1:13 p.m.
Hi Kumar,

On Tue,  5 May 2009 07:04:55 -0500 Kumar Gala <galak@kernel.crashing.org> wrote:
>
> We shouldn't directly access sysdata to get the device node.  We should
> be calling pci_device_to_OF_node().
> 
> Signed-off-by: Kumar Gala <galak@kernel.crashing.org>

It looks ok, but all I can say for now is that it doesn not make things
worse.  During 2.6.28, the iSeries PCI code was broken.  I am working on
a fix and after that I will be able to properly test this patch.
Stephen Rothwell - May 7, 2009, 12:52 a.m.
Hi Kumar,

On Wed, 6 May 2009 23:13:38 +1000 Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>
> On Tue,  5 May 2009 07:04:55 -0500 Kumar Gala <galak@kernel.crashing.org> wrote:
> >
> > We shouldn't directly access sysdata to get the device node.  We should
> > be calling pci_device_to_OF_node().
> > 
> > Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
> 
> It looks ok, but all I can say for now is that it doesn not make things
> worse.  During 2.6.28, the iSeries PCI code was broken.  I am working on
> a fix and after that I will be able to properly test this patch.

With my fix for the old PCI breakage, I have tested your "pci" tree for
iseries_defconfig.  Built and booted with and without CONFIG_PCI.

Tested-by: Stephen Rothwell <sfr@canb.auug.og.au>
Acked-by: Stephen Rothwell <sfr@canb.auug.og.au>

My patch, however, will probably need your attention once Ben takes
it. :-)

Patch

diff --git a/arch/powerpc/platforms/iseries/pci.c b/arch/powerpc/platforms/iseries/pci.c
index 02a634f..18343c3 100644
--- a/arch/powerpc/platforms/iseries/pci.c
+++ b/arch/powerpc/platforms/iseries/pci.c
@@ -318,6 +318,7 @@  static void __init iomm_table_allocate_entry(struct pci_dev *dev, int bar_num)
 {
 	struct resource *bar_res = &dev->resource[bar_num];
 	long bar_size = pci_resource_len(dev, bar_num);
+	struct device_node *dn = pci_device_to_OF_node(dev);
 
 	/*
 	 * No space to allocate, quick exit, skip Allocation.
@@ -335,9 +336,9 @@  static void __init iomm_table_allocate_entry(struct pci_dev *dev, int bar_num)
 	 * Allocate the number of table entries needed for BAR.
 	 */
 	while (bar_size > 0 ) {
-		iomm_table[current_iomm_table_entry] = dev->sysdata;
+		iomm_table[current_iomm_table_entry] = dn;
 		ds_addr_table[current_iomm_table_entry] =
-			iseries_ds_addr(dev->sysdata) | (bar_num << 24);
+			iseries_ds_addr(dn) | (bar_num << 24);
 		bar_size -= IOMM_TABLE_ENTRY_SIZE;
 		++current_iomm_table_entry;
 	}
@@ -410,7 +411,7 @@  void __init iSeries_pcibios_fixup_resources(struct pci_dev *pdev)
 	struct device_node *node;
 	int i;
 
-	node = find_device_node(bus, pdev->devfn);
+	node = pci_device_to_OF_node(pdev);
 	pr_debug("PCI: iSeries %s, pdev %p, node %p\n",
 		 pci_name(pdev), pdev, node);
 	if (!node) {
@@ -441,7 +442,6 @@  void __init iSeries_pcibios_fixup_resources(struct pci_dev *pdev)
 		}
 	}
 
-	pdev->sysdata = node;
 	allocate_device_bars(pdev);
 	iseries_device_information(pdev, bus, *sub_bus);
 	iommu_devnode_init_iSeries(pdev, node);