[PATCH] PCI: Expand documentation for pci_add_dma_alias()

[Logan Gunthorpe <logang@deltatee.com>]

Seeing there's been some confusion about the use of pci_add_dma_alias(),
expand the comment to describe why it must be called early and how
early it must be called.

Also, expand on the purpose of this function and common reasons it would
be used.

Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Doug Meyer <dmeyer@gigaio.com>
---
 drivers/pci/pci.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index dbfe7c4f3776..2a410745a95a 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -5395,8 +5395,19 @@ int pci_set_vga_state(struct pci_dev *dev, bool decode,
  * @dev: the PCI device for which alias is added
  * @devfn: alias slot and function
  *
- * This helper encodes 8-bit devfn as bit number in dma_alias_mask.
- * It should be called early, preferably as PCI fixup header quirk.
+ * This helper encodes an 8-bit devfn as a bit number in dma_alias_mask
+ * which is used to program permissible BDF source addresses for
+ * DMA requests in an IOMMU. This is useful for non-transparent bridges
+ * which use BDFs that are not visible on the bus as proxies for devices
+ * in another domain. It's also useful for grouping of multiple devices
+ * that must be done without a host endpoint driver.
+ *
+ * This function must be called early, before the IOMMU creates
+ * the groups for the device which happens with a BUS_NOTIFY_ADD_DEVICE
+ * event. This occurs before the driver's probe function is called seeing
+ * the groups must be ready before any DMA mappings for the device are created.
+ * Therefore, this function cannot be called in a driver and, instead,
+ * should be called as a PCI fixup header quirk.
  */
 void pci_add_dma_alias(struct pci_dev *dev, u8 devfn)
 {
-- 
2.11.0