On Fri, 6 Apr 2012, Xi Wang wrote: > On Apr 6, 2012, at 5:18 PM, Greg Kroah-Hartman wrote: > > How does that avoid the overflow? We still would have allocated a huge > > chunk, which would not be good. How about bounding the size of nents > > instead? > > kmalloc_array() checks for integer overflow. I agree with you that it's > better to have a tight limit. We might also need to bound iterations to > avoid overflow: context.count = param->sglen * param->iterations. > > How about limit them to 1024? That number seems much too small. People run usbtest for days at a time. Admittedly, they probably run a bunch of relatively short tests over and over again, but we shouldn't rule out the possibility of running a long test. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
