Hi, On Fri, Dec 25, 2009 at 12:02:06PM -0700, Steve Brown wrote: > On 12/23/2009 01:57 AM, Andreas Mohr wrote: >>>> And I would be very pleased if ehci-ssb.c could find its way into mainline >>>> pretty soon, >>>> >>> Yeah. Care to do so? >>> >> Huh what me? :) >> >> >> I'll get this nailed. I wouldn't want this unfortunate thing to happen again. >> >> Thanks, >> >> Andreas Mohr >> > I had planned to further investigate the problem reported at > https://forum.openwrt.org/viewtopic.php?id=16999. Although, pl2303 seems > to be working for you.
I'm truly sorry, but no, while it is fully working USB-wise, I did not actually test any serial connection! What works nicely for me is my FTDI-based device (8U232AM: i.e. full-speed!!), but I didn't have any serial test device to use with my other FTDI and PL2303 serial connector adapters. But I can offer to bring along some serial mouse or so and actively test things on this router (which currently is on the remote side of things but fortunately happily chugging along, so no dice at the moment). Specific to this problem report (which I already got to know rather intimately before, thank you very much :-P): I'm not sure whether they actively used an external self-powered high-speed hub. The SMSC 2502 is suspicious (pretty strongly worded deprecation warning on SMSC site, and somewhere it was claimed that power is quite weak, causing USB errors, especially when using both ports of the builtin hub). I'm now always using an external "industrial" 7-port hub on one builtin port _only_, and loving it (recently I added all USB gadgets I could find nearby, thus plugged in and out maybe 8 devices such as storage/serial/audio/bluetooth/hid, with total power consumption of 7.4W via 1 central power supply). Current setup is router/7-port hub/USB stick/FTDI home automation with exactly 5.0W. But OTOH the WL-520GU has one port only, thus people have to use a hub for anything useful (as also reported in the 2009-08-06 posting). BTW, I _never_ (to my knowledge) actually got an UE (Unrecoverable Error) in HC registers (and dmesg!) on my implementation of the BCM5354 USB HC, just other USB-related issues. BTW, IIRC my BCM5354 (WL-500gP v2) has a ~ 0813 marking. It certainly was quite a bit earlier than the HK0829 marking in the thread, sadly. And in my case it curiously wasn't prefixed HK (Hong Kong?) but something else. > I should have realized that the code would get out of sync and taken > mb's advice to send it upstream earlier. > > Thanks for taking this on, Yes, I will submit quilt updates ASAP, and then also push it towards mainline. Thanks for your (and other people's) nice amount of base work in this area! Andreas Mohr _______________________________________________ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
