Messages in this thread | | | Date | Wed, 1 Aug 2018 23:11:48 -0700 | From | Dennis Zhou <> | Subject | Re: [PATCH v2] block: make iolatency avg_lat exponentially decay |
| |
Hi Tejun,
On Wed, Aug 01, 2018 at 09:10:11AM -0700, Tejun Heo wrote: > On Wed, Aug 01, 2018 at 08:49:58AM -0700, Tejun Heo wrote: > > Hello, > > > > On Tue, Jul 31, 2018 at 05:25:59PM -0700, Dennis Zhou wrote: > > ... > > > + /* > > > + * CALC_LOAD takes in a number stored in fixed point representation. > > > + * Because we are using this for IO time in ns, the values stored > > > + * are significantly larger than the FIXED_1 denominator (2048). > > > + * Therefore, rounding errors in the calculation are negligible and > > > + * can be ignored. > > > + */ > > > + exp_idx = min_t(int, BLKIOLATENCY_NR_EXP_FACTORS - 1, > > > + iolat->cur_win_nsec / BLKIOLATENCY_EXP_BUCKET_SIZE); > > > > Build bot is complaining about naked 64bit div. Should use one of the > > div64*() helpers. > > > > Looks good to me. Once Johannes's concerns are addressed, please feel > > free to add > > Ooh, one nitpick. total_lat_avg is a bit of a misnomer now. Maybe > rename to lat_avg? >
Yeah, that makes sense. I've renamed it to lat_avg and updated the Documentation file as well.
Thanks, Dennis
| |