[prev in list] [next in list] [prev in thread] [next in thread]
List: kde-panel-devel
Subject: Re: Review Request 126205: If the user runs failsafe mode, use software rendering in QtQuick
From: Martin_Gräßlin <mgraesslin () kde ! org>
Date: 2015-11-30 11:01:58
Message-ID: 20151130110158.29696.93231 () mimi ! kde ! org
[Download RAW message or body]
--===============7622394124083265785==
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
> On Nov. 30, 2015, 10:18 a.m., Martin Gräßlin wrote:
> > What does QT_XCB_FORCE_SOFTWARE_OPENGL?
>
> David Edmundson wrote:
> It makes Qt apps pretend LIBGL_ALWAYS_SOFTWARE is set.
>
> Martin Gräßlin wrote:
> Sorry, but that doesn't do what you hope it does:
>
> - only affects Mesa drivers
> - if Mesa is broken for whatever reason the software rasterizer will also be broken
> - kwin does something completely different with failsafe: no OpenGL at all.
>
> So if you get crashers due to e.g. broken NVIDIA driver, this won't fix it.
>
> David Edmundson wrote:
> Yeah, I knew it didn't fix nvidia.
>
> I know it's not perfect, but I have seen user's on bugzilla with positive results \
> from it, so when I saw this code block (whilst editing something else) it seemed \
> like it might fix some situations, and I can't see any real downsides.
> Martin Gräßlin wrote:
> > I can't see any real downsides
>
> the downside is that it switches to CPU rendering. Which seems not the best thing \
> to do in general. I cannot imagine a situation where the actual mesa drivers are \
> broken and llvmpipe wouldn't be broken, too. So I just don't see what this would \
> improve for fail safe. I don't see a possible improvment, so for me the downside \
> (CPU rendering) seems worse. Any pointers to bug reports where users had a broken \
> mesa, but a still working llvmpipe can change my mind ;-)
> David Edmundson wrote:
> The Intel fiasco (so one bug a trillion billion times)
> https://bugs.kde.org/show_bug.cgi?id=349519#c28
>
> and openGL < 2 hardware
> https://bugzilla.redhat.com/show_bug.cgi?id=1249280#c1
>
> Fedora ended up with an xinit script that sets this env var in some cases (which is \
> a bit weird given it's a condition I'm sure Qt could easily check itself) \
> http://pkgs.fedoraproject.org/cgit/qt5-qtbase.git/tree/10-qt5-check-opengl2.sh
convinced!
- Martin
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/126205/#review88943
-----------------------------------------------------------
On Nov. 30, 2015, 10:13 a.m., David Edmundson wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/126205/
> -----------------------------------------------------------
>
> (Updated Nov. 30, 2015, 10:13 a.m.)
>
>
> Review request for Plasma.
>
>
> Repository: plasma-workspace
>
>
> Description
> -------
>
> Graphics seem more likely than anything else to be broken, if the user
> explicitly selects failsafe mode from the DM we should follow the same
> rationale as kwin and not trust the user's drivers.
>
>
> Diffs
> -----
>
> startkde/startkde.cmake 41a8975cce1fb2a4e7a034e697ce6e2cc59d5b1e
>
> Diff: https://git.reviewboard.kde.org/r/126205/diff/
>
>
> Testing
> -------
>
>
> Thanks,
>
> David Edmundson
>
>
--===============7622394124083265785==
MIME-Version: 1.0
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit
<html>
<body>
<div style="font-family: Verdana, Arial, Helvetica, Sans-Serif;">
<table bgcolor="#f9f3c9" width="100%" cellpadding="12" style="border: 1px #c9c399 \
solid; border-radius: 6px; -moz-border-radius: 6px; -webkit-border-radius: 6px;"> \
<tr> <td>
This is an automatically generated e-mail. To reply, visit:
<a href="https://git.reviewboard.kde.org/r/126205/">https://git.reviewboard.kde.org/r/126205/</a>
</td>
</tr>
</table>
<br />
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: \
10px;"> <p style="margin-top: 0;">On November 30th, 2015, 10:18 a.m. CET, <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;"><p style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: inherit;">What does QT_XCB_FORCE_SOFTWARE_OPENGL?</p></pre> \
</blockquote>
<p>On November 30th, 2015, 10:25 a.m. CET, <b>David Edmundson</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;"><p style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">It \
makes Qt apps pretend LIBGL_ALWAYS_SOFTWARE is set.</p></pre> </blockquote>
<p>On November 30th, 2015, 10:33 a.m. CET, <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;"><p style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: \
inherit;">Sorry, but that doesn't do what you hope it does:</p> <ul style="padding: \
0;text-rendering: inherit;margin: 0 0 0 1em;line-height: inherit;white-space: \
normal;"> <li style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: normal;">only affects Mesa drivers</li> <li style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: normal;">if \
Mesa is broken for whatever reason the software rasterizer will also be broken</li> \
<li style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: normal;">kwin does something completely different with failsafe: \
no OpenGL at all.</li> </ul>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: inherit;">So if you get crashers due to e.g. broken NVIDIA \
driver, this won't fix it.</p></pre> </blockquote>
<p>On November 30th, 2015, 10:50 a.m. CET, <b>David Edmundson</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;"><p style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">Yeah, \
I knew it didn't fix nvidia. </p> <p style="padding: 0;text-rendering: \
inherit;margin: 0;line-height: inherit;white-space: inherit;">I know it's not \
perfect, but I have seen user's on bugzilla with positive results from it, so when I \
saw this code block (whilst editing something else) it seemed like it might fix some \
situations, and I can't see any real downsides.</p></pre> </blockquote>
<p>On November 30th, 2015, 10:55 a.m. CET, <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;"><blockquote \
style="text-rendering: inherit;padding: 0 0 0 1em;border-left: 1px solid \
#bbb;white-space: normal;margin: 0 0 0 0.5em;line-height: inherit;"> <p \
style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: \
inherit;">I can't see any real downsides</p> </blockquote>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: inherit;">the downside is that it switches to CPU rendering. \
Which seems not the best thing to do in general. I cannot imagine a situation where \
the actual mesa drivers are broken and llvmpipe wouldn't be broken, too. So I just \
don't see what this would improve for fail safe. I don't see a possible improvment, \
so for me the downside (CPU rendering) seems worse. Any pointers to bug reports where \
users had a broken mesa, but a still working llvmpipe can change my mind \
;-)</p></pre> </blockquote>
<p>On November 30th, 2015, 11:13 a.m. CET, <b>David Edmundson</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;"><p style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">The \
Intel fiasco (so one bug a trillion billion times) \
https://bugs.kde.org/show_bug.cgi?id=349519#c28</p> <p style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">and \
openGL < 2 hardware https://bugzilla.redhat.com/show_bug.cgi?id=1249280#c1</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: \
inherit;white-space: inherit;">Fedora ended up with an xinit script that sets this \
env var in some cases (which is a bit weird given it's a condition I'm sure Qt could \
easily check itself) \
http://pkgs.fedoraproject.org/cgit/qt5-qtbase.git/tree/10-qt5-check-opengl2.sh</p></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;"><p style="padding: \
0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: \
inherit;">convinced!</p></pre> <br />
<p>- Martin</p>
<br />
<p>On November 30th, 2015, 10:13 a.m. CET, David Edmundson wrote:</p>
<table bgcolor="#fefadf" width="100%" cellspacing="0" cellpadding="12" style="border: \
1px #888a85 solid; border-radius: 6px; -moz-border-radius: 6px; \
-webkit-border-radius: 6px;"> <tr>
<td>
<div>Review request for Plasma.</div>
<div>By David Edmundson.</div>
<p style="color: grey;"><i>Updated Nov. 30, 2015, 10:13 a.m.</i></p>
<div style="margin-top: 1.5em;">
<b style="color: #575012; font-size: 10pt;">Repository: </b>
plasma-workspace
</div>
<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;">Graphics seem more likely than anything else to be broken, if the user \
explicitly selects failsafe mode from the DM we should follow the same rationale as \
kwin and not trust the user's drivers.</pre> </td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Diffs</b> </h1>
<ul style="margin-left: 3em; padding-left: 0;">
<li>startkde/startkde.cmake <span style="color: \
grey">(41a8975cce1fb2a4e7a034e697ce6e2cc59d5b1e)</span></li>
</ul>
<p><a href="https://git.reviewboard.kde.org/r/126205/diff/" style="margin-left: \
3em;">View Diff</a></p>
</td>
</tr>
</table>
</div>
</body>
</html>
--===============7622394124083265785==--
[Attachment #3 (text/plain)]
_______________________________________________
Plasma-devel mailing list
Plasma-devel@kde.org
https://mail.kde.org/mailman/listinfo/plasma-devel
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic