On 1/23/20 5:49 PM, Coly Li wrote: > On 2020/1/24 2:31 上午, Jens Axboe wrote: >> On 1/23/20 10:27 AM, Coly Li wrote: >>> On 2020/1/24 1:19 上午, Michael Lyle wrote: >>>> Hi Coly and Jens-- >>>> >>>> One concern I have with this is that it's going to wear out >>>> limited-lifetime SSDs a -lot- faster. Was any thought given to making >>>> this a tunable instead of just changing the behavior? Even if we have >>>> an anecdote or two that it seems to have increased performance for >>>> some workloads, I don't expect it will have increased performance in >>>> general and it may even be costly for some workloads (it all comes >>>> down to what is more useful in the cache-- somewhat-recently readahead >>>> data, or the data that it is displacing). >>> >>> Hi Mike, >>> >>> Copied. This is good suggestion, I will do it after I back from Lunar >>> New Year vacation, and submit it with other tested patches in following >>> v5.6-rc versions. >> >> Do you want me to just drop this patch for now from the series? >> > Hi Jens, > > OK, please drop this patch from this series. I will re-submit the patch > with sysfs interface later with other patches. Sounds good, I queued up the rest for 5.6. -- Jens Axboe
