[Mono-dev] Timers using Linux monotonic clock not system time

[Alex Rønne Petersen]

Hey,

Please make patches against the 'master' branch. Bonus points if you
make a pull request on GitHub. ;)

Thanks!

Regards,
Alex

On Wed, Aug 17, 2016 at 8:53 AM, Craig McQueen
<craig.mcqueen at innerrange.com> wrote:
> I've made a patch which applies for 4.4.0.122 (which is the version I'm using with the meta-mono layer for Yocto).
>
> I assume that for best results, I should apply it to the latest of some branch in the git repository. What branch(es) have active development, and are patches accepted for? I could apply it to the latest mono-4.4.0-branch, or some other?
>
> From: Rodrigo Kumpera [mailto:kumpera at gmail.com]
> Sent: Wednesday, 17 August 2016 2:31 AM
> To: Craig McQueen <craig.mcqueen at innerrange.com>
> Cc: mono-devel-list at lists.dot.net
> Subject: Re: [Mono-dev] Timers using Linux monotonic clock not system time
>
> Yes, we'd consider such patches.
>
> On Tue, Aug 16, 2016 at 12:06 AM, Craig McQueen <craig.mcqueen at innerrange.com> wrote:
> I wrote:
>>
>> On an embedded Linux project, we found that Mono timers are disrupted by
>> a system time change. If system time is set forwards, then timers fire too
>> early. If system time is set backwards, then timers don't fire when they
>> should. We found this is known bug #3025.
>> https://bugzilla.xamarin.com/show_bug.cgi?id=3025
>>
>> So we're using the patch from that bug. But, can this patch please go into the
>> main code? This bug seems to be a serious one as far as potential failure
>> modes (yet hard to catch in testing because it is a corner-case), so an
>> important one to fix.
>
> I need to upgrade to Mono 4.4.0 to fix another issue I reported (SMTP with SSL/TLS and intermediate certificates). I see that Mono 4.4.0 is susceptible to timers failing when system time changes. But the patch I was using for Mono 4.2.1 doesn't apply for Mono 4.4.0. So I'll have to rework the patch.
>
> I can provide an updated patch when I've figured it out. Would a patch be likely to be accepted for the mainline Mono code?
>
> --
> Craig McQueen
>
> _______________________________________________
> Mono-devel-list mailing list
> Mono-devel-list at lists.dot.net
> http://lists.dot.net/mailman/listinfo/mono-devel-list
>
>
> _______________________________________________
> Mono-devel-list mailing list
> Mono-devel-list at lists.dot.net
> http://lists.dot.net/mailman/listinfo/mono-devel-list
>