On 12/6/22 11:15?AM, Gulam Mohamed wrote: > Use ktime to change the granularity of IO accounting in block layer from > milli-seconds to nano-seconds to get the proper latency values for the > devices whose latency is in micro-seconds. After changing the granularity > to nano-seconds the iostat command, which was showing incorrect values for > %util, is now showing correct values. > > We did not work on the patch to drop the logic for > STAT_PRECISE_TIMESTAMPS yet. Will do it if this patch is ok. > > The iostat command was run after starting the fio with following command > on an NVME disk. For the same fio command, the iostat %util was showing > ~100% for the disks whose latencies are in the range of microseconds. > With the kernel changes (granularity to nano-seconds), the %util was > showing correct values. Following are the details of the test and their > output: As mentioned, this will most likely have a substantial performance impact. I'd test it, but your patch is nowhere near applying to the current block tree. Please resend it against for-6.2/block so it can get tested. -- Jens Axboe
