This patch backports the addition of ax8817x support to the usbnet driver. It also includes the ethool_ops support.
I basically like this (I read the patch rather than applying it and reading the result though) except for this #if 0" bit. Looks like the same code is present, after the minidriver bind() method, so that ought to be just a small bit of cleanup.
Plus, without additional build changes, "asx8817x.c" will conflict with "usbnet.c" in much the same way that "CDCEther.c" does.
Do you have any opinions about how different the 2.4 and 2.6 versions of this driver really need to be? I still don't trust the urb queueing support much on 2.4 (based on 2.4.10 testing that showed it was unreliable on at least the UHCIs), but that's just a #define.
Assuming this ax8817x/ethtool patch is merged (with minor changes) the "big" areas where the drivers differ right now are in support for CDC Ethernet (where 2.4 is still confused by Zaurii) and the kbuild/Config.in support.
- Dave
@@ -2221,7 +2481,9 @@
net->watchdog_timeo = TX_TIMEOUT_JIFFIES;
net->tx_timeout = usbnet_tx_timeout;
net->do_ioctl = usbnet_ioctl;
+ net->ethtool_ops = &usbnet_ethtool_ops;
+#if 0
// get rx/tx params from descriptors; avoid compiled-in details
if (!info->in || !info->out) {
int status = get_endpoints (dev,
@@ -2236,6 +2498,24 @@
dev->out = usb_sndbulkpipe (udev, info->out);
dev->maxpacket = info->epsize;
}
+#endif
+ // allow device-specific bind/init procedures
+ // NOTE net->name still not usable ...
+ if (info->bind) {
+ status = info->bind (dev, udev);
...
------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
