Re: [PATCH] net: usb: ax88179_178a: fix link status when link is set to down/up
From: Jose Ignacio Tornos Martinez <hidden>
Date: 2024-06-13 09:46:00
Also in:
linux-usb, lkml, stable
Subsystem:
networking drivers, the rest, usb networking drivers · Maintainers:
Andrew Lunn, "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds
Hello Yongqin, After some research and testing, it seems to happen that if initialization is slower the second reset from open is needed too. So, I have been working with some reproducers and I think I have the solution for detecting when there is a problem. If you can test it in your real environment that would be great. Here the patch on the latest version of the file: $ git diff drivers/net/usb/ax88179_178a.c
diff --git a/drivers/net/usb/ax88179_178a.c b/drivers/net/usb/ax88179_178a.c
index 51c295e1e823..60357796be99 100644
--- a/drivers/net/usb/ax88179_178a.c
+++ b/drivers/net/usb/ax88179_178a.c@@ -174,7 +174,6 @@ struct ax88179_data { u32 wol_supported; u32 wolopts; u8 disconnecting; - u8 initialized; }; struct ax88179_int_data {
@@ -327,7 +326,8 @@ static void ax88179_status(struct usbnet *dev, struct urb *urb) if (netif_carrier_ok(dev->net) != link) { usbnet_link_change(dev, link, 1); - netdev_info(dev->net, "ax88179 - Link status is: %d\n", link); + if (!link)
diff --git a/drivers/net/usb/ax88179_178a.c b/drivers/net/usb/ax88179_178a.c
index 51c295e1e823..60357796be99 100644
--- a/drivers/net/usb/ax88179_178a.c
+++ b/drivers/net/usb/ax88179_178a.c@@ -174,7 +174,6 @@ struct ax88179_data { u32 wol_supported; u32 wolopts; u8 disconnecting; - u8 initialized; }; struct ax88179_int_data {
@@ -327,7 +326,8 @@ static void ax88179_status(struct usbnet *dev, struct urb *urb) if (netif_carrier_ok(dev->net) != link) { usbnet_link_change(dev, link, 1); - netdev_info(dev->net, "ax88179 - Link status is: %d\n", link); + if (!link) + netdev_info(dev->net, "ax88179 - Link status is: %d\n", link); } }
@@ -1543,6 +1543,7 @@ static int ax88179_link_reset(struct usbnet *dev) GMII_PHY_PHYSR, 2, &tmp16); if (!(tmp16 & GMII_PHY_PHYSR_LINK)) { + netdev_info(dev->net, "ax88179 - Link status is: 0\n"); return 0; } else if (GMII_PHY_PHYSR_GIGA == (tmp16 & GMII_PHY_PHYSR_SMASK)) { mode |= AX_MEDIUM_GIGAMODE | AX_MEDIUM_EN_125MHZ;
@@ -1580,6 +1581,8 @@ static int ax88179_link_reset(struct usbnet *dev) netif_carrier_on(dev->net); + netdev_info(dev->net, "ax88179 - Link status is: 1\n"); + return 0; }
@@ -1678,12 +1681,21 @@ static int ax88179_reset(struct usbnet *dev) static int ax88179_net_reset(struct usbnet *dev) { - struct ax88179_data *ax179_data = dev->driver_priv; + u16 tmp16; - if (ax179_data->initialized) + ax88179_read_cmd(dev, AX_ACCESS_PHY, AX88179_PHY_ID, GMII_PHY_PHYSR, + 2, &tmp16); + if (tmp16) { + ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_MEDIUM_STATUS_MODE, + 2, 2, &tmp16); + if (!(tmp16 & AX_MEDIUM_RECEIVE_EN)) { + tmp16 |= AX_MEDIUM_RECEIVE_EN; + ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_MEDIUM_STATUS_MODE, + 2, 2, &tmp16); + } + } else { ax88179_reset(dev); - else - ax179_data->initialized = 1; + } return 0; }
In addition, I have fixed the logs to show the link correclty. If this is ok, I will submit the patch. Thanks Best regards José Ignacio