[prev in list] [next in list] [prev in thread] [next in thread]
List: kde-release-team
Subject: Re: Review Request: Manage m_preview->deco pointer around plugin juggling
From: Martin_Gräßlin <kde () martin-graesslin ! com>
Date: 2012-07-25 19:43:36
Message-ID: 20120725194336.11068.44725 () vidsolbach ! de
[Download RAW message or body]
[Attachment #2 (multipart/alternative)]
> On July 25, 2012, 2:30 p.m., Martin Gräßlin wrote:
> > I would appreciate if this would go into 4.9.0.
> >
> > Short summary for the release team:
> > * third party decorations (namely QtCurve and keramik) implemented an internal \
> > and unstable API for window decorations
> > * this API is called KDecoration*Unstable* to prevent anyone having the awesome \
> > idea to implement it
> > * we ensured that KWin and Systemsettings do not crash when an incompatible deco \
> > is found
> > * for Systemsettings there was a small bug which might lead to a crash (see \
> > referenced bug)
> > * I would like to not ship with a known crasher
> > * I hope I correctly summarized it :-)
>
> Thomas Lübking wrote:
> No worry, i'm back in civilization and will have usable access to my system in time \
> ;-) Also, despite what Ben said, 4.9 is not yet frozen - so we *should* still be \
> permitted to just push?!
> About the crasher fixed here - API safety is not related. Was the general "make it \
> snappier" commit.
> I introduced it by attempting to stop (long time present) leaking plugins (and \
> unloading them as well) Ultimately this (and similar usage) was very likey the \
> source of the GHNS crashes as well, the preview/plugin link is not robust in this \
> regard.
> Albert Astals Cid wrote:
> Yes, 4.9 is still not frozen, freeze starts at 23:59 UTC today (in around 9 hours)
Let me just point out that the freezes are confusing. I re-read the mail before \
writing it and completely mixed up the date. Can we please work on that for 4.10 and \
yes I volunteer to write any git hooks to properly freeze the branches (though I \
would prefer someone competent working on that).
- Martin
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/105721/#review16378
-----------------------------------------------------------
On July 25, 2012, 2:30 p.m., Thomas Lübking wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/105721/
> -----------------------------------------------------------
>
> (Updated July 25, 2012, 2:30 p.m.)
>
>
> Review request for kwin, Release Team and Martin Gräßlin.
>
>
> Description
> -------
>
> it is mandatory to manage loadPlugin() and destroyPreviousPlugin() using \
> disablePreview() loadPlugin() moves the present factory pointer to "old_fact" which \
> is then deleted by the succeeding destroyPreviousPlugin()
> I also left a warning into the sources for our heirs ;-)
>
>
> This addresses bug 304026.
> http://bugs.kde.org/show_bug.cgi?id=304026
>
>
> Diffs
> -----
>
> kwin/kcmkwin/kwindecoration/decorationmodel.cpp 3d3bb86
> kwin/kcmkwin/kwindecoration/preview.h 20ee869
> kwin/kcmkwin/kwindecoration/preview.cpp fe802c7
>
> Diff: http://git.reviewboard.kde.org/r/105721/diff/
>
>
> Testing
> -------
>
> Yes, quite some.
> I'm pretty sure this is it and that a sanitation on recreate is not necessary for \
> commented reasons. The actual troublemaker was the BorderSizesRole implementation
>
>
> /me wonders whether we can now also shortcut if (lib->loaded()) - gonna try that \
> tonight.
> I *may* be not be back in time (but i doubt so), so iff i've not acted on call \
> until 22:00 CEST just push this and the other one or two reviews in a row before \
> the freeze on my behalf. Thanks.
>
> Thanks,
>
> Thomas Lübking
>
>
[Attachment #5 (text/html)]
<html>
<body>
<div style="font-family: Verdana, Arial, Helvetica, Sans-Serif;">
<table bgcolor="#f9f3c9" width="100%" cellpadding="8" style="border: 1px #c9c399 \
solid;"> <tr>
<td>
This is an automatically generated e-mail. To reply, visit:
<a href="http://git.reviewboard.kde.org/r/105721/">http://git.reviewboard.kde.org/r/105721/</a>
</td>
</tr>
</table>
<br />
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: \
10px;"> <p style="margin-top: 0;">On July 25th, 2012, 2:30 p.m., <b>Martin \
Gräßlin</b> wrote:</p> <blockquote style="margin-left: 1em; border-left: 2px solid \
#d0d0d0; padding-left: 10px;"> <pre style="white-space: pre-wrap; white-space: \
-moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: \
break-word;">I would appreciate if this would go into 4.9.0.
Short summary for the release team:
* third party decorations (namely QtCurve and keramik) implemented an internal and \
unstable API for window decorations
* this API is called KDecoration*Unstable* to prevent anyone having the awesome idea \
to implement it
* we ensured that KWin and Systemsettings do not crash when an incompatible deco is \
found
* for Systemsettings there was a small bug which might lead to a crash (see \
referenced bug)
* I would like to not ship with a known crasher
* I hope I correctly summarized it :-)</pre>
</blockquote>
<p>On July 25th, 2012, 4:30 p.m., <b>Thomas Lübking</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: \
10px;"> <pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: \
-pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">No worry, i'm back \
in civilization and will have usable access to my system in time ;-) Also, despite \
what Ben said, 4.9 is not yet frozen - so we *should* still be permitted to just \
push?!
About the crasher fixed here - API safety is not related. Was the general "make \
it snappier" commit.
I introduced it by attempting to stop (long time present) leaking plugins (and \
unloading them as well) Ultimately this (and similar usage) was very likey the source \
of the GHNS crashes as well, the preview/plugin link is not robust in this \
regard.</pre> </blockquote>
<p>On July 25th, 2012, 4:57 p.m., <b>Albert Astals Cid</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: \
10px;"> <pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: \
-pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">Yes, 4.9 is still not \
frozen, freeze starts at 23:59 UTC today (in around 9 hours)</pre> </blockquote>
</blockquote>
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: \
-pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">Let me just point out \
that the freezes are confusing. I re-read the mail before writing it and completely \
mixed up the date. Can we please work on that for 4.10 and yes I volunteer to write \
any git hooks to properly freeze the branches (though I would prefer someone \
competent working on that).</pre> <br />
<p>- Martin</p>
<br />
<p>On July 25th, 2012, 2:30 p.m., Thomas Lübking wrote:</p>
<table bgcolor="#fefadf" width="100%" cellspacing="0" cellpadding="8" \
style="background-image: \
url('http://git.reviewboard.kde.org/media/rb/images/review_request_box_top_bg.png'); \
background-position: left top; background-repeat: repeat-x; border: 1px black \
solid;"> <tr>
<td>
<div>Review request for kwin, Release Team and Martin Gräßlin.</div>
<div>By Thomas Lübking.</div>
<p style="color: grey;"><i>Updated July 25, 2012, 2:30 p.m.</i></p>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Description </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" \
style="border: 1px solid #b8b5a0"> <tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: \
-moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: \
break-word;">it is mandatory to manage loadPlugin() and destroyPreviousPlugin() using \
disablePreview() loadPlugin() moves the present factory pointer to \
"old_fact" which is then deleted by the succeeding destroyPreviousPlugin()
I also left a warning into the sources for our heirs ;-)</pre>
</td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Testing </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" style="border: \
1px solid #b8b5a0"> <tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: \
-moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: \
break-word;">Yes, quite some. I'm pretty sure this is it and that a sanitation on \
recreate is not necessary for commented reasons. The actual troublemaker was the \
BorderSizesRole implementation
/me wonders whether we can now also shortcut if (lib->loaded()) - gonna try that \
tonight.
I *may* be not be back in time (but i doubt so), so iff i've not acted on call \
until 22:00 CEST just push this and the other one or two reviews in a row before the \
freeze on my behalf. Thanks.</pre> </td>
</tr>
</table>
<div style="margin-top: 1.5em;">
<b style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Bugs: </b>
<a href="http://bugs.kde.org/show_bug.cgi?id=304026">304026</a>
</div>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Diffs</b> </h1>
<ul style="margin-left: 3em; padding-left: 0;">
<li>kwin/kcmkwin/kwindecoration/decorationmodel.cpp <span style="color: \
grey">(3d3bb86)</span></li>
<li>kwin/kcmkwin/kwindecoration/preview.h <span style="color: \
grey">(20ee869)</span></li>
<li>kwin/kcmkwin/kwindecoration/preview.cpp <span style="color: \
grey">(fe802c7)</span></li>
</ul>
<p><a href="http://git.reviewboard.kde.org/r/105721/diff/" style="margin-left: \
3em;">View Diff</a></p>
</td>
</tr>
</table>
</div>
</body>
</html>
_______________________________________________
release-team mailing list
release-team@kde.org
https://mail.kde.org/mailman/listinfo/release-team
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic