[Intel-gfx] [PATCH 03/10] drm/armada: Drop fb gamma_set/get functions
Daniel Vetter
daniel at ffwll.ch
Wed Mar 30 15:29:12 UTC 2016
On Wed, Mar 30, 2016 at 01:56:06PM +0100, Russell King - ARM Linux wrote:
> On Wed, Mar 30, 2016 at 01:19:21PM +0200, Daniel Vetter wrote:
> > On Wed, Mar 30, 2016 at 1:09 PM, Russell King - ARM Linux
> > <linux at arm.linux.org.uk> wrote:
> > > On Wed, Mar 30, 2016 at 11:51:18AM +0200, Daniel Vetter wrote:
> > >> The fb helper private gamma_set/get functions are only required when
> > >> the driver supports paletted 8bit mode with fbdev. Armada uses 32bpp
> > >> unconditionally, so this is just dead code. It also doesn't do
> > >> anything really. Let's just remove it.
> > >
> > > This comment is misleading: Armada supports 16bpp formats as well as
> > > 32bpp, and the hardware does have 8bpp modes to, but I've chosen not
> > > to support the 8bpp modes.
> >
> > This is purely about the fbdev emulation (and yeah need to clarify
> > that), not about kms support in general. And these two gamma_set/get
> > hooks are _only_ used by the fbdev emulation, and only needed if you
> > do 8bit paletted mode. Ok if I change the commit message to "Armada
> > used 32bpp unconditionally for fbdev emulation, ..."?
>
> I still don't know where you get that from - the armada fbdev code
> supports more than just 32bpp. Please explain.
Ok, I lost myself in the code and missed the cmdline parsing of fbdev
depth. So seems indeed possible to create a 8bit fbdev on armada (since
armada_framebuffer_create won't reject it, and I didn't see anything
else). I'll drop this patch. Thanks for the feedback.
-Daniel
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
More information about the Intel-gfx
mailing list