On Mon, 17 Sep 2012, Henrik Rydberg wrote:
> The current patch does exactly the same sleeps, the only difference is > that the test is also done before the first sleep. Thus, the increased > delay, if any, comes from the sleep range. My understanding is that the original patch resulted in trying a max udelay(16). The new one does usleep_range(16, 32). That's a minimum 16 max 32 - so the fact that it stops read failures means the added delay is present and does help. > > > The MBP10,1 experiences a lot of write errors with this patch. Well the send_byte changes are not necessary for fixing my problem - I only converted it because of usleep_range() primarily. What happens if you drop the send_byte() part of the patch and test with only the wait_read() changes? I would be very surprised if that caused write or read failures. Parag -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
