Received-SPF: pass (google.com: domain of linux-kernel+bounces-99121-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Date: Mon, 11 Mar 2024 16:16:40 +0100
From: Johan Hovold <johan@kernel.org>
To: regressions@lists.linux.dev
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com>,
	Rob Clark <robdclark@gmail.com>,
	Dmitry Baryshkov <dmitry.baryshkov@linaro.org>,
	Kuogee Hsieh <quic_khsieh@quicinc.com>,
	Daniel Thompson <daniel.thompson@linaro.org>,
	Bjorn Andersson <quic_bjorande@quicinc.com>,
	dri-devel@lists.freedesktop.org, freedreno@lists.freedesktop.org,
	linux-arm-msm@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: drm/msm: DisplayPort hard-reset on hotplug regression in 6.8-rc1
Message-ID: <Ze8gWHK2ipXIHRAP@hovoldconsulting.com>
References: <Zd3kvD02Qvsh2Sid@hovoldconsulting.com>
 <ZesH21DcfOldRD9g@hovoldconsulting.com>
 <56de6cfb-fe0f-de30-d4d0-03c0fbb0afbb@quicinc.com>
 <ZeyOmJLlBbwnmaJN@hovoldconsulting.com>
 <Ze8Ke_M2xHyPYCu-@hovoldconsulting.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Ze8Ke_M2xHyPYCu-@hovoldconsulting.com>

On Mon, Mar 11, 2024 at 02:43:24PM +0100, Johan Hovold wrote:

> So, while it may still be theoretically possible to hit the resets after
> the revert, the HPD notify revert effectively "fixed" the regression in
> 6.8-rc1 by removing the preconditions that now made us hit it (i.e. the
> half-initialised bridge).
> 
> It seems the hotplug state machine needs to be reworked completely, but
> at least we're roughly back where we were with 6.7 (including that the
> bus clocks will never be turned of because of the rpm leaks on
> disconnect).

#regzbot introduced: e467e0bde881
#regzbot fix: 664bad6af3cb