Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 4 Apr 2019 09:09:46 +0200
From:   Gerd Hoffmann <kraxel@redhat.com>
To:     Daniel Vetter <daniel@ffwll.ch>
Cc:     David Airlie <airlied@redhat.com>,
        dri-devel <dri-devel@lists.freedesktop.org>,
        David Airlie <airlied@linux.ie>,
        open list <linux-kernel@vger.kernel.org>,
        "open list:DRM DRIVER FOR QEMU'S CIRRUS DEVICE" 
        <virtualization@lists.linux-foundation.org>
Subject: Re: [PATCH] drm/cirrus: rewrite and modernize driver.
Message-ID: <20190404070946.la52tm5rpgtxoov3@sirius.home.kraxel.org>
References: <20190403072318.31507-1-kraxel@redhat.com>
 <CAMwc25o=DZSVE5Qf5mow6es06Gk3=+6+8offXPY-JPgUkG_Shw@mail.gmail.com>
 <20190404055056.ddc2bdgjbgjj7tby@sirius.home.kraxel.org>
 <CAKMK7uFbLcziLXrwds9QZBBPFiBDC_O+X9fiuutNjsutaPW+sQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAKMK7uFbLcziLXrwds9QZBBPFiBDC_O+X9fiuutNjsutaPW+sQ@mail.gmail.com>
User-Agent: NeoMutt/20180716
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

  Hi,

> > Speaking of wayland:  Seems at least gnome-shell insists on using XR24.
> 
> Yeah XR24 is pretty much mandatory. Noralf added a few helpers to
> convert XR24 to other formats, for display not supporting anything
> else. Because userspace.

Ah, right, that is an option too.  Given we blit any display updates
anyway we could easily convert XR24 to RG24 while doing so. Guess that
is better ...

> > Well, I can reintroduce the module option.  I don't see any other
> > reasonable way to support 32bpp.  If the driver reports XR24 as
> > supported and also adds the higher resolutions (which work with RG16
> > only) to the mode list userspace will of course try the higher
> > resolutions with XR24 and struggle ...
> 
> Maybe atomic userspace is better (it should be, it can do TEST_ONLY),
> but I'm not so sure that exposing all modes for atomic clients would
> work. Also, currently not possible with our probe helpers (we don't
> refilter the list per client).

.. than this mess.

cheers,
  Gerd