Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 8BIT
In-Reply-To: <20200403091430.GA3845@duo.ucw.cz>
References: <20200402213253.GA2691@duo.ucw.cz> <20200402213506.GA2767@duo.ucw.cz> <20200403073720.GA23229@duo.ucw.cz> <20200403091430.GA3845@duo.ucw.cz>
Subject: Re: [Intel-gfx] 5.7-rc0: hangs while attempting to run X
From:   Chris Wilson <chris@chris-wilson.co.uk>
Cc:     tglx@linutronix.de, mingo@redhat.com, bp@alien8.de, hpa@zytor.com
To:     Pavel Machek <pavel@ucw.cz>, intel-gfx@lists.freedesktop.org,
        jani.nikula@linux.intel.com, joonas.lahtinen@linux.intel.com,
        kernel list <linux-kernel@vger.kernel.org>,
        rodrigo.vivi@intel.com
Message-ID: <158590540341.5852.3933814252282882685@build.alporthouse.com>
User-Agent: alot/0.8.1
Date:   Fri, 03 Apr 2020 10:16:43 +0100
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Quoting Pavel Machek (2020-04-03 10:14:30)
> Hi!
> 
> > > > Hardware is thinkpad x220. I had this crash few days ago. And today I
> > > > have similar-looking one, with slightly newer kernel. (Will post
> > > > as a follow-up).
> > 
> > As part of quest for working system, I tried 5.7-rc0, based on
> > 
> > Merge: 50a5de895dbe b4d8ddf8356d
> > Author: Linus Torvalds <torvalds@linux-foundation.org>
> > Date:   Wed Apr 1 18:18:18 2020 -0700
> > 
> > It hangs in userspace, at a time when X should be starting, and I'm
> > looking at blinking cursor.
> > 
> > 5.6-rcs worked, I'll test 5.6-final.
> 
> 5.6-final works.
> 
> Hmm...
> 
> commit f365ab31efacb70bed1e821f7435626e0b2528a6
> Merge: 4646de87d325 59e7a8cc2dcf
> Author: Linus Torvalds <torvalds@linux-foundation.org>
> Date:   Wed Apr 1 15:24:20 2020 -0700
> 
>     Merge tag 'drm-next-2020-04-01' of git://anongit.freedesktop.org/drm/drm
> 
> Let me test 4646de87d32526ee87b46c2e0130413367fb5362...that one works.
> 
> Ok, so obviously... I should
> test... f365ab31efacb70bed1e821f7435626e0b2528a6
> 
> Now, this is anti-social:
> 
> Busywait for request completion limit (ns)
> (DRM_I915_MAX_REQUEST_BUSYWAIT) [8000] (NEW)
> 
> How should I know what to answer here (or the others)
> 
> Interval between heartbeat pulses (ms) (DRM_I915_HEARTBEAT_INTERVAL) [2500] 2500
> Preempt timeout (ms, jiffy granularity) (DRM_I915_PREEMPT_TIMEOUT) [640] 640
> 
> I just took the defaults.. but... 

It is _deep_ under EXPERT.
-Chris