mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Benjamin Mahler <bmah...@apache.org>
Subject Re: Review Request 73137: Avoid using the libev default loop.
Date Wed, 20 Jan 2021 21:28:09 GMT


> On Jan. 13, 2021, 7:10 p.m., Ilya Pronin wrote:
> > 3rdparty/libprocess/src/posix/libev/libev.cpp
> > Lines 58 (patched)
> > <https://reviews.apache.org/r/73137/diff/1/?file=2244169#file2244169line58>
> >
> >     May be worth providing a default `false` value for `triggered` instead of relying
on `Latch` alawys being created using value initialization.

Yes, good point!


> On Jan. 13, 2021, 7:10 p.m., Ilya Pronin wrote:
> > 3rdparty/libprocess/src/posix/libev/libev.cpp
> > Lines 200 (patched)
> > <https://reviews.apache.org/r/73137/diff/1/?file=2244169#file2244169line212>
> >
> >     Seems unnecessary.

Yeah, I just wanted to only have two states:

(1) nullptr
(2) set to something non-null and **valid**

and avoid the third state:

(3) set to something non-null but deleted

Since these live beyond the call to stop() finishing


- Benjamin


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/73137/#review222464
-----------------------------------------------------------


On Jan. 12, 2021, 7:22 p.m., Benjamin Mahler wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/73137/
> -----------------------------------------------------------
> 
> (Updated Jan. 12, 2021, 7:22 p.m.)
> 
> 
> Review request for mesos and Ilya Pronin.
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> The default loop in libev supports child process events, which we
> don't want and explicitly ignore. We can just use dynamic loops
> instead which allows us to properly destroy the loop when libprocess
> finalizes.
> 
> This also helps minimize the changes needed once we add support for
> multiple loops, since those need to be dynamic loops as well.
> 
> 
> Diffs
> -----
> 
>   3rdparty/libprocess/src/posix/libev/libev.cpp b38e7a0f882a8c24950bdc6fd74a4d25fc68549e

> 
> 
> Diff: https://reviews.apache.org/r/73137/diff/1/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Benjamin Mahler
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message