linux-kernel - Re: [PATCHv1 03/14] drm/omap: plane: update fifo size on ovl setup

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20171201122502.ynl3aubhsoiatfby@earth>
Date:   Fri, 1 Dec 2017 13:25:02 +0100
From:   Sebastian Reichel <sebastian.reichel@...labora.co.uk>
To:     Tomi Valkeinen <tomi.valkeinen@...com>
Cc:     Tony Lindgren <tony@...mide.com>,
        Laurent Pinchart <laurent.pinchart@...asonboard.com>,
        dri-devel@...ts.freedesktop.org, linux-omap@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCHv1 03/14] drm/omap: plane: update fifo size on ovl setup

Hi Tomi,

On Fri, Dec 01, 2017 at 02:10:42PM +0200, Tomi Valkeinen wrote:
> 
> On 24/07/17 20:33, Sebastian Reichel wrote:
> > This is a workaround for a hardware bug occuring on OMAP3
> > with manually updated panels. Details about the HW bug are
> > unknown to me, but without this fix the panel refresh does
> > not work at all on Nokia N950.
> > 
> > Signed-off-by: Sebastian Reichel <sebastian.reichel@...labora.co.uk>
> > ---
> >  drivers/gpu/drm/omapdrm/dss/dispc.c | 16 ++++++++++++++++
> >  1 file changed, 16 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/omapdrm/dss/dispc.c b/drivers/gpu/drm/omapdrm/dss/dispc.c
> > index fd7504b37e3b..b6dca5ee34d4 100644
> > --- a/drivers/gpu/drm/omapdrm/dss/dispc.c
> > +++ b/drivers/gpu/drm/omapdrm/dss/dispc.c
> > @@ -1398,6 +1398,18 @@ void dispc_ovl_compute_fifo_thresholds(enum omap_plane_id plane,
> >  	}
> >  }
> >  
> > +void dispc_ovl_set_manual_fifo_threshold(enum omap_plane_id plane)
> > +{
> > +	u32 fifo_low, fifo_high;
> > +	bool use_fifo_merge = false;
> > +	bool use_manual_update = true;
> > +
> > +	dispc_ovl_compute_fifo_thresholds(plane, &fifo_low, &fifo_high,
> > +					  use_fifo_merge, use_manual_update);
> > +
> > +	dispc_ovl_set_fifo_threshold(plane, fifo_low, fifo_high);
> > +}
> > +
> >  static void dispc_ovl_set_mflag(enum omap_plane_id plane, bool enable)
> >  {
> >  	int bit;
> > @@ -2566,6 +2578,10 @@ static int dispc_ovl_setup(enum omap_plane_id plane,
> >  		oi->zorder, oi->pre_mult_alpha, oi->global_alpha,
> >  		oi->rotation_type, replication, vm, mem_to_mem);
> >  
> > +	/* manual mode needs other fifo thresholds */
> > +	if (mgr_fld_read(channel, DISPC_MGR_FLD_STALLMODE))
> > +		dispc_ovl_set_manual_fifo_threshold(plane);
> > +
> >  	return r;
> >  }
> 
> We don't have the stallmode for all channels. And on OMAP2 it's for
> RFBI, and I don't know if that needs different thresholds or not. Also,
> I'm not sure if OMAP4-5 DSI needs this. And I think it's good to mention
> the unknown bug in the comment too. And yes, it's unknown to me too, I
> never figured it out. These thresholds just seemed to work, while the
> default did not.
> 
> So I think we need to check that the platform is OMAP3 and channel is
> LCD. We can later extend that to cover OMAP4-5 if needed.

Ok. I will have a look at this later today. FWIW I test this
patchset with OMAP3 (N950) and OMAP4 (Droid 4). OMAP4 works
without this patch, but also with the changed thresholds.

-- Sebastian

Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)