< Alan Stern stern@xxxxxxxxxxxxxxxxxxx >wrote: >I have been on vacation and so have not had a chance to review this >carefully. However there are a few things about it which could be >improved. For example, the comments aren't really accurate. The >"three strikes" policy is applied automatically by the hardware; hence >the policy added by this patch should be called something else. > >Also, it's not a good idea to rely on a simple retry counter. Instead >of trying three times, it would be better to keep trying for some fixed >length of time, like 100 ms. I understood Alan's idea as follows. We don't know the timing of retrying, thus the fixed length of time is better than the counter of retrying to adjust the timing. I didn't consciously know that I adjust the timing. If my understanding is correct, I think that it's need to allow for the length of time, too. When the XactErr occurs, the error recovery succeed by the first retry on my system. I think that one-time retrying is enough. But I don't know whether a quick retry is correct. I change one-time retry from three times. I want to subsequently mull over the timing of retrying (a quick retry or a fix time retry). best regards Koichiro Saito -- 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
