As per the PCIe base spec r5.0, section 5.2, Link Down event can happen
under any of the following circumstances:
1. Fundamental/Hot reset
2. Link disable transmission by upstream component
3. Moving from L2/L3 to L0
When the event happens, the EPC driver capable of detecting it may pass the
notification to the EPF driver through link_down() callback in 'struct
pci_epc_event_ops'.
While the PCIe spec has not defined the actual behavior of the endpoint
when the Link Down event happens, we may assume that at least the ongoing
transactions need to be stopped as the link won't be active. So let's
cancel the command handler work in the callback implementation
pci_epf_test_link_down(). The work will be started again in
pci_epf_test_link_up() once the link comes back again.
Reviewed-by: Niklas Cassel <cassel@xxxxxxxxxx>
Tested-by: Niklas Cassel <cassel@xxxxxxxxxx>
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@xxxxxxxxxx>
---
drivers/pci/endpoint/functions/pci-epf-test.c | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/drivers/pci/endpoint/functions/pci-epf-test.c b/drivers/pci/endpoint/functions/pci-epf-test.c
index c8d0c51ae329..afb28df174c3 100644
--- a/drivers/pci/endpoint/functions/pci-epf-test.c
+++ b/drivers/pci/endpoint/functions/pci-epf-test.c
@@ -809,10 +809,20 @@ static int pci_epf_test_link_up(struct pci_epf *epf)
return 0;
}
+static int pci_epf_test_link_down(struct pci_epf *epf)
+{
+ struct pci_epf_test *epf_test = epf_get_drvdata(epf);
+
+ cancel_delayed_work_sync(&epf_test->cmd_handler);
+
+ return 0;
+}
+
static const struct pci_epc_event_ops pci_epf_test_event_ops = {
.epc_init = pci_epf_test_epc_init,
.epc_deinit = pci_epf_test_epc_deinit,
.link_up = pci_epf_test_link_up,
+ .link_down = pci_epf_test_link_down,
};
static int pci_epf_test_alloc_space(struct pci_epf *epf)
--
2.25.1
