PCI: Clarify the "scan anyway" comment in pci_scan_bridge()
authorAndreas Noever <andreas.noever@gmail.com>
Thu, 23 Jan 2014 20:59:22 +0000 (21:59 +0100)
committerBjorn Helgaas <bhelgaas@google.com>
Tue, 11 Feb 2014 00:11:15 +0000 (17:11 -0700)
Initially when we encountered a bus that was already present we skipped
it. Since 74710ded8e16 'PCI: always scan child buses' we continue
scanning in order to allow user triggered rescans of already existing
busses.

The old comment suggested that the reason for continuing the scan is a
bug in the i450NX chipset. This is not the case.

Signed-off-by: Andreas Noever <andreas.noever@gmail.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
drivers/pci/probe.c

index f340c947d8cbe2bc1714d2534f00558f515084bc..511a8f6d76363cf599551e678f711f919967859f 100644 (file)
@@ -805,11 +805,10 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max, int pass)
                        goto out;
 
                /*
-                * If we already got to this bus through a different bridge,
-                * don't re-add it. This can happen with the i450NX chipset.
-                *
-                * However, we continue to descend down the hierarchy and
-                * scan remaining child buses.
+                * The bus might already exist for two reasons: Either we are
+                * rescanning the bus or the bus is reachable through more than
+                * one bridge. The second case can happen with the i450NX
+                * chipset.
                 */
                child = pci_find_bus(pci_domain_nr(bus), secondary);
                if (!child) {