Re: Regression: 2.6.34 boot fails on E5405 system, bisected: de08e2c26

From: Ben Greear
Date: Wed Jul 14 2010 - 15:27:55 EST


On 07/14/2010 11:52 AM, H. Peter Anvin wrote:
So I suggest the following changes...

On 07/13/2010 07:19 PM, Jesse Barnes wrote:

static int fixed_bar_cap(struct pci_bus *bus, unsigned int devfn)
{
int pos;
u32 pcie_cap = 0, cap_data;
printk("fixed_bar_cap, bus: %p devfn: %u\n", bus, devfn);
pos = PCIE_CAP_OFFSET;
while (pos) {

while (pos>= PCIE_CAP_OFFSET) {

printk("Before read..\n");
if (raw_pci_ext_ops->read(pci_domain_nr(bus), bus->number,
devfn, pos, 4,&pcie_cap))
return 0;
printk("pcie_cap: %u", pcie_cap);

- if (pcie_cap == 0xffffffff)
- return 0;

+ if (PCI_EXT_CAP_ID(pcie_cap) == 0x0000 ||
+ PCI_EXT_CAP_ID(pcie_cap) == 0xffff)
+ break;

This seems to work for me. I'm using this patch against
2.6.34.y:

diff --git a/arch/x86/pci/mrst.c b/arch/x86/pci/mrst.c
index 1cdc02c..ee93fd0 100644
--- a/arch/x86/pci/mrst.c
+++ b/arch/x86/pci/mrst.c
@@ -61,13 +61,14 @@ static int fixed_bar_cap(struct pci_bus *bus, unsigned int devfn)
if (!raw_pci_ext_ops)
return 0;

- while (pos) {
+ while (pos >= PCIE_CAP_OFFSET) {
if (raw_pci_ext_ops->read(pci_domain_nr(bus), bus->number,
devfn, pos, 4, &pcie_cap))
return 0;

- if (pcie_cap == 0xffffffff)
- return 0;
+ if (PCI_EXT_CAP_ID(pcie_cap) == 0x0000 ||
+ PCI_EXT_CAP_ID(pcie_cap) == 0xffff)
+ break;

if (PCI_EXT_CAP_ID(pcie_cap) == PCI_EXT_CAP_ID_VNDR) {
raw_pci_ext_ops->read(pci_domain_nr(bus), bus->number,
@@ -76,7 +77,7 @@ static int fixed_bar_cap(struct pci_bus *bus, unsigned int devfn)
return pos;
}

- pos = pcie_cap >> 20;
+ pos = (pcie_cap >> 20) & 0xffc;
}

return 0;


Thanks,
Ben



printk("Checking vendor..\n");

if (PCI_EXT_CAP_ID(pcie_cap) == PCI_EXT_CAP_ID_VNDR) {
printk("reading domain_nr\n");
raw_pci_ext_ops->read(pci_domain_nr(bus), bus->number,
devfn, pos + 4, 4,&cap_data);
printk("cap_data: %u\n", cap_data);
if ((cap_data& 0xffff) == PCIE_VNDR_CAP_ID_FIXED_BAR)
return pos;
}

pos = pcie_cap>> 20;

pos = (pcie_cap>> 20)& 0xffc;

printk("pos after shift: %i\n", pos);
}

printk("Returning from fixed_bar_cap\n");
return 0;
}



I thought a related bug was fixed already; the code should be returning
all zeros for non-existent BAR reads.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/


--
Ben Greear <greearb@xxxxxxxxxxxxxxx>
Candela Technologies Inc http://www.candelatech.com

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/