From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon@freedesktop.org Subject: [Bug 105145] vaExportSurfaceHandle interaction with surface interlaced flag prevents switching on vaapi deinterlacing dynamically Date: Mon, 11 Jun 2018 13:09:00 +0000 Message-ID: References: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0374187627==" Return-path: Received: from culpepper.freedesktop.org (culpepper.freedesktop.org [IPv6:2610:10:20:722:a800:ff:fe98:4b55]) by gabe.freedesktop.org (Postfix) with ESMTP id 8827789AD2 for ; Mon, 11 Jun 2018 13:09:00 +0000 (UTC) In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: dri-devel@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org --===============0374187627== Content-Type: multipart/alternative; boundary="15287225402.CE1CcA.19628" Content-Transfer-Encoding: 7bit --15287225402.CE1CcA.19628 Date: Mon, 11 Jun 2018 13:09:00 +0000 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated https://bugs.freedesktop.org/show_bug.cgi?id=3D105145 --- Comment #12 from Christian K=C3=B6nig --- (In reply to k.philipp from comment #11) > Is the interlaced format necessary in the first place? Unfortunately yes it is. The pixel shader used for de-interlacing needs to have access to the top/bo= ttom field separately and that only works efficiently when they aren't mangled together. We could try to double the stride to work around that, but that i= sn't easily doable with the interface we have and we would stress the memory interface quite a bit more. Additional to that most hardware de-interlacer work only with that layout, = .e.g it is actually the more natural one for MPEG2. And that's also the reason w= hy for example NVidia hardware only works with that as well. That HEVC and VP9 only support progressive layout in the output format is a= lso only logical because those formats don't support interlaced content. --=20 You are receiving this mail because: You are the assignee for the bug.= --15287225402.CE1CcA.19628 Date: Mon, 11 Jun 2018 13:09:00 +0000 MIME-Version: 1.0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated

Comme= nt # 12 on bug 10514= 5 from Christian K=C3= =B6nig 
(In reply to k.philipp from comment #11)
> Is the interlaced format necessary in the first =
place?

Unfortunately yes it is.

The pixel shader used for de-interlacing needs to have access to the top/bo=
ttom
field separately and that only works efficiently when they aren't mangled
together. We could try to double the stride to work around that, but that i=
sn't
easily doable with the interface we have and we would stress the memory
interface quite a bit more.

Additional to that most hardware de-interlacer work only with that layout, =
.e.g
it is actually the more natural one for MPEG2. And that's also the reason w=
hy
for example NVidia hardware only works with that as well.

That HEVC and VP9 only support progressive layout in the output format is a=
lso
only logical because those formats don't support interlaced content.

You are receiving this mail because:
  • You are the assignee for the bug.
= --15287225402.CE1CcA.19628-- --===============0374187627== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KZHJpLWRldmVs IG1haWxpbmcgbGlzdApkcmktZGV2ZWxAbGlzdHMuZnJlZWRlc2t0b3Aub3JnCmh0dHBzOi8vbGlz dHMuZnJlZWRlc2t0b3Aub3JnL21haWxtYW4vbGlzdGluZm8vZHJpLWRldmVsCg== --===============0374187627==--