qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Philippe Mathieu-Daudé" <philmd@redhat.com>
To: "Alex Bennée" <alex.bennee@linaro.org>,
	"Peter Maydell" <peter.maydell@linaro.org>
Cc: QEMU Developers <qemu-devel@nongnu.org>
Subject: Re: [PATCH] tests/acceptance/boot_linux: Skip slow Aarch64 'virt' machine TCG test
Date: Fri, 8 May 2020 15:07:04 +0200	[thread overview]
Message-ID: <15b8c809-d8f8-4ca0-9815-d35a87fc5f83@redhat.com> (raw)
In-Reply-To: <871rnv5vva.fsf@linaro.org>

On 5/7/20 10:32 PM, Alex Bennée wrote:
> 
> Peter Maydell <peter.maydell@linaro.org> writes:
> 
>> On Thu, 7 May 2020 at 19:38, Alex Bennée <alex.bennee@linaro.org> wrote:
>>>
>>>
>>> Philippe Mathieu-Daudé <philmd@redhat.com> writes:
>>>
>>>> The BootLinuxAarch64.test_virt_tcg is reported to take >7min to run.
>>>> Add a possibility to users to skip this particular test, by setting
>>>> the AVOCADO_SKIP_SLOW_TESTS environment variable:
>>>>
>>>>    $ AVOCADO_SKIP_SLOW_TESTS=please make check-acceptance
>>>>    ...
>>>>      (05/88) tests/acceptance/boot_linux.py:BootLinuxAarch64.test_virt_tcg: SKIP: Test takes >7min
>>>>    ...
>>>
>>> 7m = 420s - I'm seeing:
>>>
>>>    hyperfine "./tests/venv/bin/avocado run ./tests/acceptance/boot_linux.py:BootLinuxAarch64.test_virt_tcg"
>>>    Benchmark #1: ./tests/venv/bin/avocado run ./tests/acceptance/boot_linux.py:BootLinuxAarch64.test_virt_tcg
>>>      Time (mean ± σ):     162.179 s ±  3.138 s    [User: 204.726 s, System: 9.663 s]
>>>      Range (min … max):   158.651 s … 170.036 s    10 runs
>>>
>>> Is this on very slow hardware?
>>
>> Intel(R) Xeon(R) W-2145 CPU @ 3.70GHz 16-core
>> clang sanitizer build, which probably slows it down a bit.
>>
>> But even 200 seconds is an order of magnitude slower than any
>> of the other tests that check-acceptance runs. I think we
>> should be aiming for tests here to be ~30 seconds at most,
>> or the whole thing will take forever by the time we have
>> decent coverage of most machines.
> 
> I think we could say the same for a whole bunch of the tests that are
> based on full distro downloads. What is this one running under the hood
> anyway?

Download time is now counted separately, this was one big feature 
request after the last KVM forum.

The second request was to improve test timing management, like having 
different timers tracking if the test is behaving as expected. I.e. you 
might want to boot a VM and run 'ping -c 3'. The whole test shouldn't 
take more than 7min, but we want to be sure the 'ping' command doesn't 
take more than 3.x seconds. I haven't heard about a such feature ready yet.

I really hope you are using a common cache location and now download the 
artifacts each time...

> 
> 
>>
>> thanks
>> -- PMM
> 
> 



  reply	other threads:[~2020-05-08 13:18 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-07 16:22 [PATCH] tests/acceptance/boot_linux: Skip slow Aarch64 'virt' machine TCG test Philippe Mathieu-Daudé
2020-05-07 18:38 ` Alex Bennée
2020-05-07 18:45   ` Peter Maydell
2020-05-07 20:32     ` Alex Bennée
2020-05-08 13:07       ` Philippe Mathieu-Daudé [this message]
2020-05-09 13:18 ` Lukas Straub
2020-05-11  9:11   ` Peter Maydell
2020-05-11 16:51     ` Lukas Straub
2020-05-26  9:14       ` Philippe Mathieu-Daudé

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=15b8c809-d8f8-4ca0-9815-d35a87fc5f83@redhat.com \
    --to=philmd@redhat.com \
    --cc=alex.bennee@linaro.org \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).