Thread (18 messages) 18 messages, 5 authors, 2025-01-03

RE: [EXTERNAL] [PATCH] net: phy: don't issue a module request if a driver is available

From: Suman Ghosh <hidden>
Date: 2025-01-03 07:48:17

quoted
quoted
	mutex_init(&dev->lock);
	INIT_DELAYED_WORK(&dev->state_queue, phy_state_machine);

-	/* Request the appropriate module unconditionally; don't
-	 * bother trying to do so only if it isn't already loaded,
-	 * because that gets complicated. A hotplug event would have
-	 * done an unconditional modprobe anyway.
-	 * We don't do normal hotplug because it won't work for MDIO
+	/* We don't do normal hotplug because it won't work for MDIO
	 * -- because it relies on the device staying around for long
	 * enough for the driver to get loaded. With MDIO, the NIC
	 * driver will get bored and give up as soon as it finds that @@ -
724,7 +745,8 @@ struct phy_device *phy_device_create(struct mii_bus
*bus, int addr, u32 phy_id,
		int i;

		for (i = 1; i < num_ids; i++) {
-			if (c45_ids->device_ids[i] == 0xffffffff)
+			if (c45_ids->device_ids[i] == 0xffffffff ||
+			    phy_driver_exists(c45_ids->device_ids[i]))
				continue;

			ret = phy_request_driver_module(dev, @@ -732,7 +754,7 @@
struct
quoted
quoted
phy_device *phy_device_create(struct mii_bus *bus, int addr, u32
phy_id,
			if (ret)
				break;
		}
-	} else {
+	} else if (!phy_driver_exists(phy_id)) {
[Suman] Can we add this phy_driver_exists() API call before the
if/else check?
quoted
Not really, as in case of C45 PHYs we have to check for drivers using
(multiple) IDs, which are different from phy_id.
[Suman] Hmm got it. We can isolate the else part but that won't make much difference. Thank you!!
  
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help