@@ -648,6 +648,12 @@ static int ax88178_reset(struct usbnet *dev)asix_sw_reset(dev,0);msleep(150);+/* LyconSys FiberGecko: Turn on PHY */+if(dev->driver_info->flags==0x20061201){+/* Power on Phy: set gpio2=high */+asix_write_gpio(dev,AX_GPIO_GPO2EN|AX_GPIO_GPO_2,50);+}+asix_sw_reset(dev,AX_SWRESET_PRL|AX_SWRESET_IPPD);msleep(150);
@@ -946,6 +952,19 @@ static const struct driver_info hg20f9_info = {.data=FLAG_EEPROM_MAC,};+staticconststructdriver_infolyconsys_fibergecko100_info={+.description="LyconSys FiberGecko 100 USB 2.0 to Ethernet Adapter",+.bind=ax88178_bind,+.status=asix_status,+.link_reset=ax88178_link_reset,+.reset=ax88178_link_reset,+.flags=FLAG_ETHER|FLAG_FRAMING_AX|FLAG_LINK_INTR|+FLAG_MULTI_PACKET,+.rx_fixup=asix_rx_fixup_common,+.tx_fixup=asix_tx_fixup,+.data=0x20061201,+};+staticconststructusb_device_idproducts[]={{// Linksys USB200M
From: Max Schulze <hidden> Date: 2025-01-26 11:42:28
Signed-off-by: Max Schulze <redacted>
Tested-by: Max Schulze <redacted>
Suggested-by: David Hollis <redacted>
Reported-by: Sven Kreiensen <redacted>
Link: https://marc.info/?l=linux-netdev&m=140742722321328
---
This patch had previously been suggested at "Link:".
However, I found that the flag quirk is not necessary and I suspect
it has never worked (because it references ".flag" whereas the
identifying value is in ".data")
I have compiled this and tested successfully with two devices.
As it now only adds the USB Id it generates no extra maintenance
burden that's why I suggest it for inclusion.
drivers/net/usb/asix_devices.c | 17 +++++++++++++++++
1 file changed, 17 insertions(+)
From: Max Schulze <hidden> Date: 2025-01-26 12:12:46
Signed-off-by: Max Schulze <redacted>
Tested-by: Max Schulze <redacted>
Suggested-by: David Hollis <redacted>
Reported-by: Sven Kreiensen <redacted>
---
v2: change Spacing on Initializer, change Mailing List link
This patch had previously been suggested at
https://lore.kernel.org/netdev/1407426826-11335-2-git-send-email-dhollis@davehollis.com/
However, I found that the flag quirk is not necessary and I suspect
it has never worked (because it references ".flag" whereas the
identifying value is in ".data")
I have compiled this and tested successfully with two devices.
As it now only adds the USB Id it generates no extra maintenance
burden that's why I suggest it for inclusion.
drivers/net/usb/asix_devices.c | 17 +++++++++++++++++
1 file changed, 17 insertions(+)
From: Andrew Lunn <andrew@lunn.ch> Date: 2025-01-26 16:33:33
For obvious reasons I can't take patches without any changelog messages,
but maybe other subsystems have more relaxed rules :(
Same for netdev, we like to see a commit message.
If the intention is this is to be merged via netdev, please read:
https://www.kernel.org/doc/html/latest/process/maintainer-netdev.html
Also not that netdev is closed at the moment due to the emerge window.
Andrew
---
pw-bot: cr
From: Max Schulze <hidden> Date: 2025-02-12 15:12:01
The FiberGecko is a small USB module that connects a 100 Mbit/s SFP
Signed-off-by: Max Schulze <redacted>
Tested-by: Max Schulze <redacted>
Suggested-by: David Hollis <redacted>
Reported-by: Sven Kreiensen <redacted>
---
v3: resend out of merge-window, with commit message
v2: change Spacing on Initializer, change Mailing List link
This patch had previously been suggested at
https://lore.kernel.org/netdev/1407426826-11335-2-git-send-email-dhollis@davehollis.com/
However, I found that the flag quirk is not necessary and I suspect
it has never worked (because it references ".flag" whereas the
identifying value is in ".data")
I have compiled this and tested successfully with two devices.
As it now only adds the USB Id it generates no extra maintenance
burden that's why I suggest it for inclusion.
drivers/net/usb/asix_devices.c | 17 +++++++++++++++++
1 file changed, 17 insertions(+)