[lkml]   [2021]   [Jan]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH] target/file: don't zero iter before iov_iter_bvec
On 11/01/2021 02:06, Chaitanya Kulkarni wrote:
> On 1/9/21 13:29, Pavel Begunkov wrote:
>> On 09/01/2021 20:52, Chaitanya Kulkarni wrote:
>>> On 1/9/21 12:40, Pavel Begunkov wrote:
>>>> I expect you won't find any, but such little things can pile up
>>>> into a not-easy-to-spot overhead over time.
>>> That is what I suspected with the resulting assembly. The commit log
>>> needs to document that there is no direct impact on the performance
>> It's obvious that 3-4 extra mov $0 off(%reg) won't change performance
>> but still hasn't been formally confirmed ...
> This is obvious for you and me since we spent time into looking into
> resulting assembly not every reviewer is expected to do that see [1].
>>> which can be seen with this patch, but this is nice to have
>> ... so if you don't mind, I won't be resending just for that.
> As per commit log guidelines [1] you have to quantify the optimization.
> Since you cannot quantify the optimization modify the commit log explaining

And then you see "Optimizations usually aren’t free but trade-offs
between", and the patch doesn't fall under it.

Let me be frank, I see it more like as a whim. If the maintainer agrees
with that strange requirement of yours and want to bury it under
bureaucracy, fine by me, don't take it, I don't care, but I haven't
ever been asked here to do that for patches as this.

> that there is not significant performance benefit observe.

It's not "I cannot" but rather "I haven't even tried to and expect...".
Don't mix, there is a huge difference between.

Pavel Begunkov

 \ /
  Last update: 2021-01-11 03:35    [W:0.077 / U:0.016 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site