From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Hutchings Subject: Re: [PATCH 4.4 23/31] drm/bridge: adv7511: Use work_struct to defer hotplug handing to out of irq context Date: Fri, 13 Oct 2017 14:53:49 +0100 Message-ID: <1507902829.22379.12.camel@codethink.co.uk> References: <20170912165247.436880944@linuxfoundation.org> <20170912165248.541459149@linuxfoundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8bit Return-path: In-Reply-To: <20170912165248.541459149@linuxfoundation.org> Sender: linux-kernel-owner@vger.kernel.org To: John Stultz , Laurent Pinchart Cc: stable@vger.kernel.org, David Airlie , Archit Taneja , Wolfram Sang , Lars-Peter Clausen , dri-devel@lists.freedesktop.org, Thong Ho , Nhan Nguyen , Greg Kroah-Hartman , LKML List-Id: dri-devel@lists.freedesktop.org On Tue, 2017-09-12 at 09:56 -0700, Greg Kroah-Hartman wrote: > 4.4-stable review patch.  If anyone has any objections, please let me > know. > > ------------------ > > From: John Stultz > > commit 518cb7057a59b9441336d2e88a396d52b6ab0cce upstream. > > I was recently seeing issues with EDID probing, where > the logic to wait for the EDID read bit to be set by the > IRQ wasn't happening and the code would time out and fail. > > Digging deeper, I found this was due to the fact that > IRQs were disabled as we were running in IRQ context from > the HPD signal. > > Thus this patch changes the logic to handle the HPD signal > via a work_struct so we can be out of irq context. [...] Shouldn't there also be a cancel_work_sync() in the remove function? Ben. -- Ben Hutchings Software Developer, Codethink Ltd.