linux-kernel - Re: [PATCH] drm/sun4i: uncouple DSI dotclock divider from TCON0_DCLK

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: <2219295.iZASKD2KPV@jernej-laptop>
Date:   Sat, 08 Apr 2023 09:07:37 +0200
From:   Jernej Škrabec <jernej.skrabec@...il.com>
To:     Maxime Ripard <maxime@...no.tech>,
        Roman Beranek <romanberanek@...oud.com>
Cc:     Frank Oltmanns <frank@...manns.dev>, Chen-Yu Tsai <wens@...e.org>,
        David Airlie <airlied@...il.com>,
        Daniel Vetter <daniel@...ll.ch>,
        Samuel Holland <samuel@...lland.org>,
        dri-devel@...ts.freedesktop.org,
        linux-arm-kernel@...ts.infradead.org, linux-sunxi@...ts.linux.dev,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] drm/sun4i: uncouple DSI dotclock divider from TCON0_DCLK_REG

Dne sreda, 05. april 2023 ob 14:34:11 CEST je Roman Beranek napisal(a):
> Hello Maxime,
> 
> On Wed Mar 29, 2023 at 9:58 PM CEST, Maxime Ripard wrote:
> > > In order to preserve semantic correctness however, I propose to preface
> > > the change with a patch that renames sun4i_dotclock and tcon-pixel-clock
> > > such that dot/pixel is replaced with d/data. What do you think?
> > 
> > I don't think it's exposed to the userspace in any way so it makes sense
> > to me
> Here's a new series that includes those renames:
> <https://lore.kernel.org/all/20230331110245.43527-1-me@crly.cz/>
> 
> It turns out however that the new dclk rates can't be set exactly as
> requested without touching pll-video0*, tcon0 now therefore gets
> reparented from pll-mipi to pll-video0-2x which, as it further turns
> out, breaks DSI. While simply forbidding the video0-2x mux option seems
> to me as the right way to go because there's not much use for it with
> non-DSI interfaces either besides the opportunity to power pll-mipi
> down, I'd like to run by you first.

It's been a long time since I looked at A64 HDMI clocks, but IIRC, pll-video0 
is the only useful source for HDMI PHY (as opposed to HDMI controller.)
So question remains how to properly support both displays at the same time.

Have you ever tried to make HDMI and DSI work at the same time? This is one of 
issues of the PinePhone IIUC.


> 
> Kind regards,
> Roman
> 
> * As pll-mipi doesn't have CLK_SET_RATE_PARENT flag set, pll-video0
>   retains its boot-time rate of 294 MHz set by sunxi-dw-hdmi driver
>   in u-boot. Why 294 MHz (as opposed to the default rate of 297 MHz)?
>   The driver actually asks for 297 MHz, clock_set_pll3 rounds it to
>   294 MHz though because it limits itself to 6 MHz steps.

Yeah, we added CLK_SET_RATE_PARENT flag to several clocks after initial driver 
was merged. Adding this flag sounds completely reasonable.

Best regards,
Jernej