From kde-panel-devel  Tue Oct 31 19:35:25 2017
From: =?utf-8?q?Martin_Fl=C3=B6ser?= <noreply () phabricator ! kde ! org>
Date: Tue, 31 Oct 2017 19:35:25 +0000
To: kde-panel-devel
Subject: D8582: [platforms/drm] Add support for rotation property on the Plane
Message-Id: <20171031193524.104659.079290B1215EFB11 () phabricator ! kde ! org>
X-MARC-Message: https://marc.info/?l=kde-panel-devel&m=150947854126493
MIME-Version: 1
Content-Type: multipart/mixed; boundary="--ef18b7e3feb14b6c9d6306dd6bc4d483"

--ef18b7e3feb14b6c9d6306dd6bc4d483
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset="ascii"
Mime-Version: 1.0

graesslin added a comment.


  In https://phabricator.kde.org/D8582#162577, @davidedmundson wrote:
  
  > Do you think we'll need a compositor fallback?
  >
  > *If* we do have a compositor fallback, then I don't see the point of a DRM specific codepath, it's not going to have any slower having a transform in painting, and will just result in one code path not getting tested.
  >
  > To mostly answer my own question:
  >
  > In hwcomposer, it seems we have a transform attribute on the layer.
  >
  > For nested wayland we can set a transform on our wl_surface.
  >
  > For FBDev or nested in X, I don't think we can. But probably don't care?
  
  
  Yep. And that's why I don't want to add a compositor fallback. Doing the fallback looked quite complex when I looked at it due to multi screen.

REPOSITORY
  R108 KWin

REVISION DETAIL
  https://phabricator.kde.org/D8582

To: graesslin, #kwin, #plasma, subdiff
Cc: davidedmundson, plasma-devel, kwin, bwowk, ZrenBot, progwolff, lesliezhai, ali-mohamed, hardening, jensreuterberg, abetts, sebas, apol, mart

--ef18b7e3feb14b6c9d6306dd6bc4d483
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html; charset="ascii"
Mime-Version: 1.0

<table><tr><td style=3D"">graesslin added a comment.
</td><a style=3D"text-decoration: none; padding: 4px 8px; margin: 0 8px 8px=
; float: right; color: #464C5C; font-weight: bold; border-radius: 3px; back=
ground-color: #F7F7F9; background-image: linear-gradient(to bottom,#fff,#f1=
f0f1); display: inline-block; border: 1px solid rgba(71,87,120,.2);" href=
=3D"https://phabricator.kde.org/D8582" rel=3D"noreferrer">View Revision</a>=
</tr></table><br /><div><div><blockquote style=3D"border-left: 3px solid #8=
C98B8;
          color: #6B748C;
          font-style: italic;
          margin: 4px 0 12px 0;
          padding: 8px 12px;
          background-color: #F8F9FC;">
<div style=3D"font-style: normal;
          padding-bottom: 4px;">In <a href=3D"https://phabricator.kde.org/D=
8582#162577" style=3D"background-color: #e7e7e7;
          border-color: #e7e7e7;
          border-radius: 3px;
          padding: 0 4px;
          font-weight: bold;
          color: black;text-decoration: none;" rel=3D"noreferrer">D8582#162=
577</a>, <a href=3D"https://phabricator.kde.org/p/davidedmundson/" style=3D"
              border-color: #f1f7ff;
              color: #19558d;
              background-color: #f1f7ff;
                border: 1px solid transparent;
                border-radius: 3px;
                font-weight: bold;
                padding: 0 4px;" rel=3D"noreferrer">@davidedmundson</a> wro=
te:</div>
<div style=3D"margin: 0;
          padding: 0;
          border: 0;
          color: rgb(107, 116, 140);"><p>Do you think we&#039;ll need a com=
positor fallback?</p>

<p>*If* we do have a compositor fallback, then I don&#039;t see the point o=
f a DRM specific codepath, it&#039;s not going to have any slower having a =
transform in painting, and will just result in one code path not getting te=
sted.</p>

<p>To mostly answer my own question:</p>

<p>In hwcomposer, it seems we have a transform attribute on the layer.</p>

<p>For nested wayland we can set a transform on our wl_surface.</p>

<p>For FBDev or nested in X, I don&#039;t think we can. But probably don&#0=
39;t care?</p></div>
</blockquote>

<p>Yep. And that&#039;s why I don&#039;t want to add a compositor fallback.=
 Doing the fallback looked quite complex when I looked at it due to multi s=
creen.</p></div></div><br /><div><strong>REPOSITORY</strong><div><div>R108 =
KWin</div></div></div><br /><div><strong>REVISION DETAIL</strong><div><a hr=
ef=3D"https://phabricator.kde.org/D8582" rel=3D"noreferrer">https://phabric=
ator.kde.org/D8582</a></div></div><br /><div><strong>To: </strong>graesslin=
, KWin, Plasma, subdiff<br /><strong>Cc: </strong>davidedmundson, plasma-de=
vel, kwin, bwowk, ZrenBot, progwolff, lesliezhai, ali-mohamed, hardening, j=
ensreuterberg, abetts, sebas, apol, mart<br /></div>
--ef18b7e3feb14b6c9d6306dd6bc4d483--