ubuntu-phone team mailing list archive

[Kevin Gunn <kevin.gunn@xxxxxxxxxxxxx>]

Just a followup on this thread. Thanks for those that gave it a shot and
provided feedback.
Results look really good.
There are some minor knock-on effect bugs - we're tracking those here
https://bugs.launchpad.net/bugs/+bugs?field.searchtext=nonblockswap
All of which have some level of debug and proposed solution.
No doubt this is an improvement even with the minor "regressions", so I
plan to move forward in a landing attempt, with the team remaining focused
to clean up the regressions.

br,kg



On Wed, Apr 16, 2014 at 5:19 PM, Kevin Gunn <kevin.gunn@xxxxxxxxxxxxx>wrote:

> Hi all,
> Some of you may have been following the various threads on the "Qt5.2
> queuing events when screen is off" bug through any of the following:
> https://bugs.launchpad.net/mir/+bug/1292306
> https://bugreports.qt-project.org/browse/QTBUG-37677
> https://lists.launchpad.net/ubuntu-phone/msg07458.html
>
> Short summary on the conclusion, we're working on altering Mir to not
> block on swapbuffers in the instance of occluded apps & screen off and
> relying on the shell to enforce the toolkit side channel to halt rendering.
>
> Given that this is a significant behavioral change we're taking a
> systematic approach and hoping to get some additional testing and
> road-milage before landing. At the moment we have branches available which
> to deliver the desired change. We welcome anyone who would like to help out
> by testing... and if you'd like to help, please follow this wiki I threw
> together.
> https://wiki.ubuntu.com/Mir/NonBlockingSwapTesting
>
> br, kg
>