PCI/IOV: Assume SR-IOV VFs support extended config space.

The SR-IOV specification requires both PFs and VFs to implement a PCIe
capability.  Generally this is sufficient to assume extended config space
is present, but we generally also perform additional tests to make sure the
extended config space is reachable and not simply an alias of standard
config space.  For a VF to exist extended config space must be accessible
on the PF, therefore we can also assume it to be accessible on the VF.
This enables a micro performance optimization previously implemented in
commit 975bb8b4dc ("PCI/IOV: Use VF0 cached config space size for other
VFs") to speed up probing of VFs.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Cc: KarimAllah Ahmed <karahmed@amazon.de>
Cc: Hao Zheng <yinhe@linux.alibaba.com>
This commit is contained in:
Alex Williamson 2019-06-13 16:57:20 -06:00 committed by Bjorn Helgaas
parent 76bf6a8634
commit 06013b647c

View File

@ -1561,6 +1561,21 @@ int pci_cfg_space_size(struct pci_dev *dev)
u32 status; u32 status;
u16 class; u16 class;
#ifdef CONFIG_PCI_IOV
/*
* Per the SR-IOV specification (rev 1.1, sec 3.5), VFs are required to
* implement a PCIe capability and therefore must implement extended
* config space. We can skip the NO_EXTCFG test below and the
* reachability/aliasing test in pci_cfg_space_size_ext() by virtue of
* the fact that the SR-IOV capability on the PF resides in extended
* config space and must be accessible and non-aliased to have enabled
* support for this VF. This is a micro performance optimization for
* systems supporting many VFs.
*/
if (dev->is_virtfn)
return PCI_CFG_SPACE_EXP_SIZE;
#endif
if (dev->bus->bus_flags & PCI_BUS_FLAGS_NO_EXTCFG) if (dev->bus->bus_flags & PCI_BUS_FLAGS_NO_EXTCFG)
return PCI_CFG_SPACE_SIZE; return PCI_CFG_SPACE_SIZE;