Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 6 Jan 2023 17:21:19 -0800
From:   Brian Norris <briannorris@chromium.org>
To:     Michel =?iso-8859-1?Q?D=E4nzer?= <michel.daenzer@mailbox.org>
Cc:     Heiko =?iso-8859-1?Q?St=FCbner?= <heiko@sntech.de>,
        Sean Paul <seanpaul@chromium.org>,
        "kernelci.org bot" <bot@kernelci.org>,
        linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
        Sandy Huang <hjc@rock-chips.com>,
        linux-rockchip@lists.infradead.org, stable@vger.kernel.org
Subject: Re: [PATCH 2/2] drm/rockchip: vop: Leave vblank enabled in
 self-refresh
Message-ID: <Y7jJDw4RcNceyLbR@google.com>
References: <20230105174001.1.I3904f697863649eb1be540ecca147a66e42bfad7@changeid>
 <20230105174001.2.Ic07cba4ab9a7bd3618a9e4258b8f92ea7d10ae5a@changeid>
 <9455bc5b-2074-4f48-71a7-5c816ee19a78@mailbox.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <9455bc5b-2074-4f48-71a7-5c816ee19a78@mailbox.org>
Precedence: bulk

On Fri, Jan 06, 2023 at 12:42:54PM +0100, Michel D?nzer wrote:
> On 1/6/23 02:40, Brian Norris wrote:
> > --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
> > +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
> > @@ -719,11 +719,11 @@ static void vop_crtc_atomic_disable(struct drm_crtc *crtc,
> >  
> >  	mutex_lock(&vop->vop_lock);
> >  
> > -	drm_crtc_vblank_off(crtc);
> > -
> >  	if (crtc->state->self_refresh_active)
> >  		goto out;
> >  
> > +	drm_crtc_vblank_off(crtc);
> > +
> >  	/*
> >  	 * Vop standby will take effect at end of current frame,
> >  	 * if dsp hold valid irq happen, it means standby complete.
> 
> The out label immediately unlocks vop->vop_lock again, seems a bit pointless. :)

Oops, of course. Will change that in v2.

> AFAICT the self_refresh_active case should just return immediately,
> the out label would also send any pending atomic commit completion
> event, which seems wrong now that the vblank interrupt is left
> enabled.

If I return immediately and drop that completion, I hit a warning:

[    4.423876] WARNING: CPU: 5 PID: 58 at drivers/gpu/drm/drm_atomic_helper.c:2460 drm_atomic_helper_commit_hw_done+0xe0/0xe8
...
[    4.424036] Call trace:
[    4.424039]  drm_atomic_helper_commit_hw_done+0xe0/0xe8
[    4.424045]  drm_atomic_helper_commit_tail_rpm+0x58/0x7c
...

I plan to leave it as-is for v2.

> (I also wonder if drm_crtc_vblank_off doesn't take care of
> that already, at least amdgpu doesn't do this explicitly AFAICT)

I'm not sure.

Brian