DPDK CI discussions
 help / color / mirror / Atom feed
From: Brandon Lo <blo@iol.unh.edu>
To: "Juraj Linkeš" <juraj.linkes@pantheon.tech>
Cc: "ci@dpdk.org" <ci@dpdk.org>,
	"Ruifeng.Wang@arm.com" <Ruifeng.Wang@arm.com>,
	"Honnappa.Nagarahalli@arm.com" <Honnappa.Nagarahalli@arm.com>
Subject: Re: [dpdk-ci] SVE Testing Temporarily Disabled
Date: Fri, 29 Oct 2021 11:17:15 -0400	[thread overview]
Message-ID: <CAOeXdvZY9sWzKz1gKjgqGdeHuX2ivUhb0zobg674Xs2YWPz5Vg@mail.gmail.com> (raw)
In-Reply-To: <c45d44de525147bda938d66053faa9fb@pantheon.tech>

Hi Juraj,

That sounds good to me. I can see how the queue plays after enabling
the sve test again with these new changes.
This will probably happen early next week, just so that I can watch
over it throughout the work week.

I will also let you know if I am going to migrate over to the
cross-compile solution in the future.

Thanks,
Brandon

On Fri, Oct 29, 2021 at 10:56 AM Juraj Linkeš
<juraj.linkes@pantheon.tech> wrote:
>
> Hi Brandon,
>
> I don't think we can get it lower using this approach, since we disabled everything we could. Can we try enabling the job again and see whether a queue builds up?
>
> I mentioned another approach in the CI call - do a cross compilation on x86 (which is fast), then use those artifacts in the emulated docker environment to run the lpm test. Lincoln mentioned you'd plan sometime in the future and this would surely be good enough. In the meantime, we could use the above approach if it's good enough, execution time-wise.
>
> Juraj
>
> > -----Original Message-----
> > From: Brandon Lo <blo@iol.unh.edu>
> > Sent: Friday, October 29, 2021 3:55 PM
> > To: Juraj Linkeš <juraj.linkes@pantheon.tech>
> > Cc: ci@dpdk.org; Ruifeng.Wang@arm.com; Honnappa.Nagarahalli@arm.com
> > Subject: Re: SVE Testing Temporarily Disabled
> >
> > Hi Juraj,
> >
> > Thanks for the suggestion. Now, all steps (compile and test) take ~15 minutes in
> > total. The original time was ~1 hour.
> > Do you believe this is good enough for now, or do you want to try to get it
> > lower?
> >
> > Thanks,
> > Brandon
> >
> > On Thu, Oct 28, 2021 at 5:14 AM Juraj Linkeš <juraj.linkes@pantheon.tech>
> > wrote:
> > >
> > > Hi Brandon,
> > >
> > > I've looked into reducing the compile time in the SVE job. We don't need to
> > build everything, just the lpm lib and hns3 driver. Could you try adding "-
> > Ddisable_libs=* -Denable_drivers=net/hns3" to and removing "-Dexamples=all"
> > from the meson compile command and test how much this helps?
> > >
> > > Thanks,
> > > Juraj
> > >
> > > > -----Original Message-----
> > > > From: Brandon Lo <blo@iol.unh.edu>
> > > > Sent: Wednesday, September 8, 2021 5:40 PM
> > > > To: Juraj Linkeš <juraj.linkes@pantheon.tech>
> > > > Cc: ci@dpdk.org; Ruifeng.Wang@arm.com;
> > Honnappa.Nagarahalli@arm.com
> > > > Subject: Re: SVE Testing Temporarily Disabled
> > > >
> > > > Hi Juraj,
> > > >
> > > > Yes, this is the emulated SVE environment within containers.
> > > > Now that the backlog has been reduced, I can look at ways to reduce
> > > > run time or prevent another backlog for when this is enabled again.
> > > >
> > > > Thanks,
> > > > Brandon
> > > >
> > > > On Fri, Sep 3, 2021 at 4:02 AM Juraj Linkeš <juraj.linkes@pantheon.tech>
> > wrote:
> > > > >
> > > > > Hi Brandon,
> > > > >
> > > > > Is this the compile + one test in an aarch64 with SVE emulated
> > environment?
> > > > >
> > > > > Thanks,
> > > > > Juraj
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Brandon Lo <blo@iol.unh.edu>
> > > > > > Sent: Thursday, September 2, 2021 7:53 PM
> > > > > > To: ci@dpdk.org; Juraj Linkeš <juraj.linkes@pantheon.tech>
> > > > > > Subject: SVE Testing Temporarily Disabled
> > > > > >
> > > > > > Hi All,
> > > > > >
> > > > > > Currently the SVE testing takes quite a bit of time to complete
> > > > > > which takes up slots from UNH.
> > > > > > We have temporarily disabled the testing in an attempt to reduce
> > > > > > the
> > > > backlog.
> > > > > > Afterwards, we will look into ways to reduce the overall time needed.
> > > > > >
> > > > > > Thanks,
> > > > > > Brandon
> > > > > >
> > > > > > --
> > > > > > Brandon Lo
> > > > > > UNH InterOperability Laboratory
> > > > > > 21 Madbury Rd, Suite 100, Durham, NH 03824 blo@iol.unh.edu
> > > > > > www.iol.unh.edu
> > > > >
> > > >
> > > >
> > > > --
> > > > Brandon Lo
> > > > UNH InterOperability Laboratory
> > > > 21 Madbury Rd, Suite 100, Durham, NH 03824 blo@iol.unh.edu
> > > > www.iol.unh.edu
> > >
> >
> >
> > --
> > Brandon Lo
> > UNH InterOperability Laboratory
> > 21 Madbury Rd, Suite 100, Durham, NH 03824 blo@iol.unh.edu
> > www.iol.unh.edu
>


-- 
Brandon Lo
UNH InterOperability Laboratory
21 Madbury Rd, Suite 100, Durham, NH 03824
blo@iol.unh.edu
www.iol.unh.edu

      reply	other threads:[~2021-10-29 15:17 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-02 17:53 Brandon Lo
2021-09-03  8:02 ` Juraj Linkeš
2021-09-08 15:40   ` Brandon Lo
2021-10-28  9:14     ` Juraj Linkeš
2021-10-29 13:55       ` Brandon Lo
2021-10-29 14:56         ` Juraj Linkeš
2021-10-29 15:17           ` Brandon Lo [this message]

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=CAOeXdvZY9sWzKz1gKjgqGdeHuX2ivUhb0zobg674Xs2YWPz5Vg@mail.gmail.com \
    --to=blo@iol.unh.edu \
    --cc=Honnappa.Nagarahalli@arm.com \
    --cc=Ruifeng.Wang@arm.com \
    --cc=ci@dpdk.org \
    --cc=juraj.linkes@pantheon.tech \
    /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).