| Hi, On Sun, 12 Jan 2003, Jochen Friedrich wrote:
| is there any particular reason why usb_string() tries to ask for the | length of the string first before actually getting the string? tbuf has a | fixed length of 256 anyways, so this just looks like an unnessessary step | to me. | | I have one particular USB device which fails miserably with EOVERFLOW in | the first query (with len=2) for a particular string. When removing the | extra step and always running the query with len=256, the device works | fine (yet another QUIRK, it seems). Quirk in this device? This device can't respond nicely to a request with len=2? It's broken then, isn't it? USB spec allows this request. Has your device passed any USB working group tests? IIRC, the code is like this because some devices don't work with the code modified as you described it. I don't recall why they didn't (2-3 years back), but they were also quirky, just in the opposite sense. At any rate, what the Linux-USB code does is allowed by USB spec. -- ~Randy ------------------------------------------------------- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
