On Thu, Aug 30, 2018 at 11:44 AM, <[email protected]> wrote: > On Wed, 2018-08-29 at 19:22 +0200, Andreas Müller wrote: >> On Wed, Aug 29, 2018 at 4:16 PM, Richard Purdie >> <[email protected]> wrote: >> > On Wed, 2018-08-29 at 00:19 +0200, Andreas Müller wrote: >> > > The patch should increase performance for libsdl2 on GLES2 too. >> > > >> > > Signed-off-by: Andreas Müller <[email protected]> >> > > --- >> > > ...01-GLES2-Get-sin-cos-out-of-vertex-shader.patch | 141 >> > > +++++++++++++++++++++ >> > > meta/recipes-graphics/libsdl2/libsdl2_2.0.8.bb | 1 + >> > > 2 files changed, 142 insertions(+) >> > > create mode 100644 meta/recipes-graphics/libsdl2/libsdl2/0001- >> > > GLES2- >> > > Get-sin-cos-out-of-vertex-shader.patch >> > > >> > > diff --git a/meta/recipes-graphics/libsdl2/libsdl2/0001-GLES2- >> > > Get- >> > > sin-cos-out-of-vertex-shader.patch b/meta/recipes- >> > > graphics/libsdl2/libsdl2/0001-GLES2-Get-sin-cos-out-of-vertex- >> > > shader.patch >> > > new file mode 100644 >> > > index 0000000000..621b7ea1a0 >> > > --- /dev/null >> > > +++ b/meta/recipes-graphics/libsdl2/libsdl2/0001-GLES2-Get-sin- >> > > cos- >> > > out-of-vertex-shader.patch >> > > @@ -0,0 +1,141 @@ >> > > +From c215ba1d52a3d4ef03af3ab1a5baa1863f812aed Mon Sep 17 >> > > 00:00:00 >> > > 2001 >> > > +From: =?UTF-8?q?Andreas=20M=C3=BCller?= <[email protected] >> > > > >> > > +Date: Fri, 24 Aug 2018 23:10:25 +0200 >> > > +Subject: [PATCH] GLES2: Get sin/cos out of vertex shader >> > > +MIME-Version: 1.0 >> > > +Content-Type: text/plain; charset=UTF-8 >> > > +Content-Transfer-Encoding: 8bit >> > > + >> > > +The only place angle is activated and causes effect is >> > > RenderCopyEx. >> > > All other >> > > +methods which use vertex shader, leave angle disabled and cause >> > > useless sin/cos >> > > +calculation in shader. >> > > + >> > > +To get around shader's interface is changed to a vector that >> > > contains results >> > > +of sin and cos. To behave properly when disabled, cos value is >> > > set >> > > with offset >> > > +-1.0 making 0.0 default when deactivated. >> > > + >> > > +As nice side effect it simplifies GLES2_UpdateVertexBuffer: All >> > > attributes are >> > > +vectors now. >> > > + >> > > +Additional background: >> > > + >> > > +* On RaspberryPi it gives a performace win for operations. >> > > Tested >> > > with >> > > + [1] numbers go down for 5-10% (not easy to estimate due to >> > > huge >> > > variation). >> > > +* SDL_RenderCopyEx was tested with [2] >> > > +* It works around left rotated display caused by low accuracy >> > > sin >> > > implemetation >> > > + in RaspberryPi/VC4 [3] >> > > + >> > > +Upstream-Status: Submitted [4] >> > >> > If I'm reading the logs correctly, this was actually merged >> > upstream so >> > the status would become Accepted or Backport? >> > >> > > +[1] https://github.com/schnitzeltony/sdl2box >> > > +[2] https://github.com/schnitzeltony/sdl2rendercopyex >> > > +[3] https://github.com/anholt/mesa/issues/110 >> > > +[4] https://hg.libsdl.org/SDL/rev/e5a666405750 >> > >> > Cheers, >> > >> > Richard >> >> LOL: V1 I tried 'Applied [4]' >> >> and received patch failure message with the following hint: >> >> > Valid status Pending, Accepted, Backport, Denied, Inappropriate >> >> [reason], Submitted [where] >> >> My favourite would have been >> >> > Accepted[4] >> >> to state out clearly that '[4]' is oe-specific and not part of the >> original patch. If I read the hint correctly that would have caused >> another patch failure. So I chose Submitted. >> >> So Accepted without []? >> >> Sorry to ask such kind of questions... > > I'd have thought Accepted should have a []. A quick grep shows other > patches which have that in the tree so I suspect that help text is > misleading... > V2 with Accepted?
Andreas -- _______________________________________________ Openembedded-core mailing list [email protected] http://lists.openembedded.org/mailman/listinfo/openembedded-core
