Hi All, The command duration is measured in sg.c using jiffies_to_msecs(jiffies). The HZ set up in my ARM system is 100, so this time measurement can only provide resolution of 10ms (or +/- 5ms). Since the AT commands in sequential read mode can be as fast as about 1ms, this inaccuracy could easily add up. Do we need to increase the time resolution to at least 100us (i.e, 1/10th of 1ms)? SG duration is measured at the sg level which is still several layers above LLD, so SG duration may not be quite close to device time spent on each command. Is there any other way to get more accurate measurement of command duration on the device? Thanks, Fajun - To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
