| Message ID | 1632445619-9211-1-git-send-email-igor.druzhinin@citrix.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | pci: fix handling of PCI bridges with subordinate bus number 0xff | expand |
On 24.09.2021 03:06, Igor Druzhinin wrote: > --- a/xen/drivers/passthrough/pci.c > +++ b/xen/drivers/passthrough/pci.c > @@ -364,7 +364,7 @@ static struct pci_dev *alloc_pdev(struct pci_seg *pseg, u8 bus, u8 devfn) > switch ( pdev->type = pdev_type(pseg->nr, bus, devfn) ) > { > u16 cap; > - u8 sec_bus, sub_bus; > + u16 sec_bus, sub_bus; > > case DEV_TYPE_PCIe2PCI_BRIDGE: > case DEV_TYPE_LEGACY_PCI_BRIDGE: First of all you want to also address the same issue in free_pdev() then. Further, since we're switching away from u16, uint16_t would be the legitimate replacement. Plus, since cap is then of the same type, fold all three variable declarations into a single line. Finally, as per ./CODING_STYLE, fixed width types should be used only where strictly needed. I can't see a reason for any of these to be other than "unsigned int". Jan
diff --git a/xen/drivers/passthrough/pci.c b/xen/drivers/passthrough/pci.c index fc4fa2e..48b415c 100644 --- a/xen/drivers/passthrough/pci.c +++ b/xen/drivers/passthrough/pci.c @@ -364,7 +364,7 @@ static struct pci_dev *alloc_pdev(struct pci_seg *pseg, u8 bus, u8 devfn) switch ( pdev->type = pdev_type(pseg->nr, bus, devfn) ) { u16 cap; - u8 sec_bus, sub_bus; + u16 sec_bus, sub_bus; case DEV_TYPE_PCIe2PCI_BRIDGE: case DEV_TYPE_LEGACY_PCI_BRIDGE:
Bus number 0xff is valid according to the PCI spec. Using u8 typed sub_bus and assigning 0xff to it will result in the following loop getting stuck. for ( ; sec_bus <= sub_bus; sec_bus++ ) {...} Just change its type to u16 the same way that is already handled in dmar_scope_add_buses(). Signed-off-by: Igor Druzhinin <igor.druzhinin@citrix.com> --- xen/drivers/passthrough/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)