Review Request 125028: Monitor for the clock changes from the kernel

[Alex Richardson]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/125028/#review84757
-----------------------------------------------------------


Looks good to me


dataengines/time/timeengine.cpp (line 32)
<https://git.reviewboard.kde.org/r/125028/#comment58665>

    #ifdef Q_OS_LINUX



dataengines/time/timeengine.cpp (line 66)
<https://git.reviewboard.kde.org/r/125028/#comment58666>

    #ifdef Q_OS_LINUX



dataengines/time/timeengine.cpp (line 67)
<https://git.reviewboard.kde.org/r/125028/#comment58663>

    We should probably add `TFD_CLOEXEC` but glibc headers don't seem to have it: Linux source has `#define TFD_CLOEXEC O_CLOEXEC` so we can use `O_CLOEXEC`



dataengines/time/timeengine.cpp (line 70)
<https://git.reviewboard.kde.org/r/125028/#comment58664>

    Maybe add a comment that glibc doesn't provide the `TFD_TIMER_CANCEL_ON_SET` constant and that's why you are using 3 instead of the constants.
    
    Also shouldn't there be an error check here?
    
    It is unlikely that someone runs with Linux < 3.0  and we therefore get a `EINVAL`, but it can still return `ENOMEN` or `EBADF` if the call to `timerfd_create` failed.


- Alex Richardson


On Sept. 2, 2015, 11:29 a.m., David Edmundson wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/125028/
> -----------------------------------------------------------
> 
> (Updated Sept. 2, 2015, 11:29 a.m.)
> 
> 
> Review request for Plasma.
> 
> 
> Repository: plasma-workspace
> 
> 
> Description
> -------
> 
> At the moment we update the clock when we resume from suspend or from
> the time KCM.
> 
> However the signal from the clock KCM isn't very good as it signals just
> after it requests an NTP update or clock change, not necessarily when
> that finishes.
> 
> Also we miss any external NTP updates which might occur at other times
> throughout the day, such as a big change when we connect to the network
> 
> This method also avoids needing solid to tell us when we're resuming
> from suspend, which is deprecated API and also seems slightly slower.
> 
> BUG: 344870
> 
> 
> Diffs
> -----
> 
>   dataengines/time/timeengine.cpp 2fd9792209ff5e78bd3dee1ed938eb0b1173de8d 
> 
> Diff: https://git.reviewboard.kde.org/r/125028/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> David Edmundson
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/plasma-devel/attachments/20150902/928fc3cb/attachment-0001.html>