Re: [PATCH v6] amba: Remove deferred device addition

From: Guenter Roeck <linux@roeck-us.net>
To: Saravana Kannan <saravanak@google.com>
Cc: Russell King <linux@armlinux.org.uk>,
	Philipp Zabel <p.zabel@pengutronix.de>,
	Rob Herring <robh@kernel.org>,
	Ulf Hansson <ulf.hansson@linaro.org>,
	Linus Walleij <linus.walleij@linaro.org>,
	Sudeep Holla <sudeep.holla@arm.com>,
	Nicolas Saenz Julienne <nsaenz@kernel.org>,
	Geert Uytterhoeven <geert+renesas@glider.be>,
	Marek Szyprowski <m.szyprowski@samsung.com>,
	Kefeng Wang <wangkefeng.wang@huawei.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	patches@armlinux.org.uk, kernel-team@android.com,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v6] amba: Remove deferred device addition
Date: Tue, 9 Aug 2022 18:34:40 -0700	[thread overview]
Message-ID: <0c6ef9a3-bbb1-9f1c-7f00-ceb05589594e@roeck-us.net> (raw)
In-Reply-To: <CAGETcx_ATE6vy9YhygHnBA2P1GDVi54np-=E+50F+cwnj6Wg4A@mail.gmail.com>

On 8/9/22 17:42, Saravana Kannan wrote:
> On Tue, Aug 9, 2022 at 3:30 AM Guenter Roeck <linux@roeck-us.net> wrote:
>>
>> Hi,
>>
>> On Wed, Jul 27, 2022 at 11:19:35AM -0700, Saravana Kannan wrote:
>>> The uevents generated for an amba device need PID and CID information
>>> that's available only when the amba device is powered on, clocked and
>>> out of reset. So, if those resources aren't available, the information
>>> can't be read to generate the uevents. To workaround this requirement,
>>> if the resources weren't available, the device addition was deferred and
>>> retried periodically.
>>>
>> ...
>>
>> This patch results in a large number of crashes in various qemu
>> emulations. Crash and bisect logs below. Reverting this patch
>> fixes the problem.
> 
> Hey Guenter,
> 
> Thanks for the report. I'm kinda surprised because I had a pl011 probe
> successfully in my local testing.
> 

Maybe it only happens with qemu emulations, or with certain configurations.

> I'm wondering if you are having an interaction with some other changes I made.
> Can you try pulling this series in and see if it helps?
> 
> https://lore.kernel.org/lkml/20220727185012.3255200-1-saravanak@google.com/
> 
>> Additional information: The decoded stack trace suggests that the
>> "id" parameter of pl011_probe() may be NULL.
> 
> That's strange! I'll take a closer look once you confirm that the
> series above doesn't help.
> 

That series does not make a difference.

Guenter