Date: Wed, 10 Jan 2018 19:34:19 +0100 (CET)
From: Thomas Gleixner <tglx@linutronix.de>
To: Meelis Roos <mroos@linux.ee>
cc: Ingo Molnar <mingo@kernel.org>,
        Linux Kernel list <linux-kernel@vger.kernel.org>,
        x86@kernel.org, Dou Liyang <douly.fnst@cn.fujitsu.com>
Subject: Re: lapic-related boot crash in 4.15-rc1
In-Reply-To: <alpine.LRH.2.21.1801101303350.11024@math.ut.ee>
Message-ID: <alpine.DEB.2.20.1801101931590.1919@nanos>
References: <alpine.LRH.2.21.1801101303350.11024@math.ut.ee>
User-Agent: Alpine 2.20 (DEB 67 2015-01-07)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org

On Wed, 10 Jan 2018, Meelis Roos wrote:
> On 3 of my test computers, boot hangs with 4.15 git kernels. So far I 
> have traced it down to 4.14.0 being good and 4.15-rc1 being bad (bisect 
> is slow because the computers are somwehat remote). Also because of 
> trying to find when it started, I have not tries newer than rc5 
> kernels.
> 
> The hang happens after Performance Events feature table has been printed 
> but before APIC initialization messages normally appear ("Enabling APIC 
> mode:  Flat.  Using 0 I/O APICs" or something like that).
> 
> It happens on at least 3 distinct computers:
> * AMD Athlon with AMD 750 chipset
> * P3 1 GHz with i815 chipset
> * P4 1.7 GHz with SiS chipset
> 
> They all have have "lapic" parameter because they were telling me that 
> lapic has been disabled by BIOS and I re-enabled it with the parameter 
> to get apic timer interrupts. This worked up until 4.14.
> 
> None of the computers have IO-APICs.

Oh. Then it possibly falls into Dou's realm.

Dou?

Thanks,

	tglx