Subject: Re: [PATCH] misc: kgdbts: Display progress of asynchronous tests
To: Jason Wessel <jason.wessel@windriver.com>
Cc: kgdb-bugreport@lists.sourceforge.net, Arnd Bergmann <arnd@arndb.de>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        linux-kernel@vger.kernel.org, patches@linaro.org
References: <20171212121036.26468-1-daniel.thompson@linaro.org>
 <31570645-24de-3625-7a43-eb6d41214478@windriver.com>
From: Daniel Thompson <daniel.thompson@linaro.org>
Message-ID: <100d1ca3-fcc3-e77b-66c4-3cd4c657e31b@linaro.org>
Date: Thu, 14 Dec 2017 16:11:39 +0000
User-Agent: Mozilla/5.0 (X11; Linux aarch64; rv:52.0) Gecko/20100101
 Thunderbird/52.5.0
MIME-Version: 1.0
In-Reply-To: <31570645-24de-3625-7a43-eb6d41214478@windriver.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1661
Lines: 38

On 14/12/17 15:20, Jason Wessel wrote:
> On 12/12/2017 06:10 AM, Daniel Thompson wrote:
>> kgdbts includes a couple of different "thrashing" style tests that
>> may have long runtimes (especially on simulated platforms) and which
>> run asynchronously. This is uncomfortable for interactive use and
>> makes setting timeouts tricky for automatic use.
> 
> 
> Do you know which test was specifically causing a problem?  It might not 
> be documented anywhere but I had usually started a user space process 
> which quickly created and deleted user space processes in order to make 
> the kgdbts tests complete quickly.

kgdbts=V1S10000 was bumping into 30 second timeouts.

IIRC this was simulating arm hardware on a fairly powerful x86 host 
machine. You can see the temporary workaround I used here:
https://github.com/daniel-thompson/kcontest/blob/master/tests/test_kgdb_selftest.py#L114

I decided this workaround was insufficient however since it would be 
rather brittle if I wanted to move to slower (more power efficient) 
hardware.


> I don't really see any issue with emitting a printk to indicate progress 
> as it is debug only and test specific.  As you know printk's change 
> timing.  If I had a dime for each time I had seen a problem go away when 
> I started adding printk's I'd have at least a 50 cents.  :-)

Agree about the interference.

I worked on the basis that these are thrashing style tests so providing 
there is a human perceivable gap between console messages (i.e. we are 
not "wasting" a CPU on an SMP system by having it in printk all the 
time) then perturbing the timing with printk() could even be beneficial.


Daniel.