On Tuesday 25 July 2006 1:06 pm, Pete Zaitcev wrote: > I find the aliasing of -110 for upper-level timeouts and the loss of > token unfortunate and pointless. I would prefer to be able to distinguish > these situation when reading user-submitted dmesg outputs. > > I looked at the way we handle those errors, and it seems like there > isn't going to be much harm from letting it to fall through on the > default handling. I only tuned HID because that thing was historically > sensitive. > > What do you think?
That change seems reasonable ... I had a similar thought back when I started rewriting OHCI for 2.6, but wasn't keen on chasing the consequences of that change on top of all the fixes for locking and race bugs. However a "grep ETIMEDOUT drivers/usb/*/*.[hc]" shows more than HID needs to change, you should update all the relevant drivers not just that one ... e.g. other input drivers and the network drivers are all pretty good about robust fault recovery and this change would cause breakage if you don't fix those too. - Dave ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
