On Tue, Aug 11, 2009 at 08:20:32AM +0200, Stephen Kitt wrote: > Hi Dmitry, > > On Mon, 10 Aug 2009 13:27:26 -0700, Dmitry Torokhov > <dmitry.torokhov@xxxxxxxxx> wrote: > > Actually I think we need to return "len" here to let userspace know how > > much data we produced. Also, could you please give the same treatment to > > JSIOCGAXMAP and JSIOCSAXMAP and I will apply it. > > Here goes! I'm not too familiar with kernel memory handling, I'll send the > fixes for the FIXMEs later on as a separate patch once I've figured things > out; I'm thinking along these lines: > * vmalloc() a buffer > * return -ENOMEM if no memory is available > * copy_from_user the data into the temporary buffer > * validate the data > * on error vfree() the buffer and return -EINVAL > * copy the data from the temporary buffer to the destination array > * vfree() the buffer > Yep, exactly, except that don't bother with vmalloc, kmalloc will do nicely since the amout of memory needed is relatively small. -- Dmitry -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
