This patch forces us to reallocate VF BARs if the totalVFs value has
increased after enabling ARI. This normally shouldn't occur, however I
have seen some non-spec devices that shift between 7 and some value greater
than 7 based on the ARI value and we want to avoid triggering any issues
with such devices.
Fixes: 3aa71da412fe ("PCI: Enable SR-IOV ARI Capable Hierarchy before reading
TotalVFs")
Signed-off-by: Alexander Duyck <adu...@mirantis.com>
---
drivers/pci/iov.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c
index 099050d78a39..238950412de0 100644
--- a/drivers/pci/iov.c
+++ b/drivers/pci/iov.c
@@ -393,7 +393,7 @@ static int sriov_init(struct pci_dev *dev, int pos)
int rc;
int nres;
u32 pgsz;
- u16 ctrl, total;
+ u16 ctrl, total, orig_total;
struct pci_sriov *iov;
struct resource *res;
struct pci_dev *pdev;
@@ -402,6 +402,7 @@ static int sriov_init(struct pci_dev *dev, int pos)
pci_pcie_type(dev) != PCI_EXP_TYPE_ENDPOINT)
return -ENODEV;
+ pci_read_config_word(dev, pos + PCI_SRIOV_TOTAL_VF, &orig_total);
pci_read_config_word(dev, pos + PCI_SRIOV_CTRL, &ctrl);
if (ctrl & PCI_SRIOV_CTRL_VFE) {
pci_write_config_word(dev, pos + PCI_SRIOV_CTRL, 0);
@@ -450,6 +451,14 @@ found:
}
iov->barsz[i] = resource_size(res);
res->end = res->start + resource_size(res) * total - 1;
+
+ /* force reallocation of BARs if total VFs increased */
+ if (orig_total < total) {
+ res->flags |= IORESOURCE_UNSET;
+ res->end -= res->start;
+ res->start = 0;
+ }
+
dev_info(&dev->dev, "VF(n) BAR%d space: %pR (contains BAR%d for
%d VFs)\n",
i, res, i, total);
i += bar64;
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
