[Barrelfish-users] Barrelfish-users Digest, Vol 25, Issue 15

Tue Feb 28 13:12:08 CET 2012

Samuel,

As Tim points out this is an exercise that I am very interested in, but resourcing it has to date been an issue. I'm not sure this would be applicable to you, but we have advertised for a HiPEAC intern to look into the issues of porting Barrelfish OS to the gem5 v7 model.

http://www.hipeac.net/phdintern/overview/8

If we do manage to fill this position I will happily coordinate with you to avoid any clash/overlap of efforts. In the interim I would be interested to know what the time constraints on your Bachelor thesis are?

Feel free to contact me directly.

Kind regards,

Matt Horsnell

> -----Original Message-----
> From: barrelfish-users-request at lists.inf.ethz.ch [mailto:barrelfish-
> users-request at lists.inf.ethz.ch]
> Sent: 28 February 2012 11:00
> To: barrelfish-users at lists.inf.ethz.ch
> Subject: Barrelfish-users Digest, Vol 25, Issue 15
>
> Send Barrelfish-users mailing list submissions to
>       barrelfish-users at lists.inf.ethz.ch
>
> To subscribe or unsubscribe via the World Wide Web, visit
>       https://lists.inf.ethz.ch/mailman/listinfo/barrelfish-users
> or, via email, send a message with subject or body 'help' to
>       barrelfish-users-request at lists.inf.ethz.ch
>
> You can reach the person managing the list at
>       barrelfish-users-owner at lists.inf.ethz.ch
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Barrelfish-users digest..."
>
>
> Today's Topics:
>
>    1. Re: building an arm kernel of barrelfish for gem5 (Samuel Hitz)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Mon, 27 Feb 2012 15:33:08 +0100
> From: Samuel Hitz <samuel.hitz at gmail.com>
> Subject: Re: [Barrelfish-users] building an arm kernel of barrelfish
>       for gem5
> To: "Tim Harris (RESEARCH)" <tharris at microsoft.com>
> Cc: "barrelfish-users at lists.inf.ethz.ch"
>       <barrelfish-users at lists.inf.ethz.ch>
> Message-ID:
>       <CA+j6bYcpuO-
> HrHo9rbVm1PVFeWt+6T9WHL2vVxM2CyjWmgWEiA at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> Hi Tim,
>
> Thanks for your advice.
>
> I'll have a closer look at the way gem5 boots the provided sample ARM
> fullsystem files. As for your second point, that's one purpose of the
> thesis to make Barrelfish support the ARMv7 gem5 simulates. I'll
> clearly
> first target a single core system, before moving on to more complex
> setups.
>
> Best,
>
> Samuel
>
> On Mon, Feb 27, 2012 at 2:55 PM, Tim Harris (RESEARCH) <
> tharris at microsoft.com> wrote:
>
> >  Hi Samuel,****
> >
> > ** **
> >
> > I remember talking about this with Matt Horsnell from ARM some time
> ago,
> > though I?m not sure if they?ve had a chance to look at BF-on-ARM
> yet.****
> >
> > ** **
> >
> > The ?molly? tool is used to wrap up the kernel and modules into a
> single
> > bootable image.  I wrote it because x64 gem5 was quite tailored to
> Linux,
> > and expected to load a single kernel passed on the simulator?s
> command
> > line.  I think that, for Linux, that kernel would proceed to load
> modules
> > from a hard disk attached to the simulated machine.  ****
> >
> > ** **
> >
> > The two initial things I?d suggest investigating are:****
> >
> > ** **
> >
> > **?         **Checking what methods gem5 supports for loading kernels
> and
> > modules on ARM.  E.g., when booting Linux, then where do things come
> from?
> >
> > ****
> >
> > **?         **Checking that the kinds of ARM system that gem5 can
> model
> > fit with the kinds of ARM system supported by BF.****
> >
> > ** **
> >
> > It may also be worth checking whether QEMU supports the same kind of
> ARM
> > system that you wish to model.  If so, then you could start off using
> QEMU
> > (for a faster debugging cycle), and then switch over to gem5 later.
> Also,
> > I?d suggest getting a simple non-heterogenous system booting and
> running
> > first, before starting on the more complex big.LITTLE setting.****
> >
> > ** **
> >
> > Thanks,****
> >
> > ** **
> >
> > Tim****
> >
> > ** **
> >
> > ** **
> >
> > ** **
> >
> > *From:* Samuel Hitz [mailto:samuel.hitz at gmail.com]
> > *Sent:* 27 February 2012 13:01
> > *To:* Tim Harris (RESEARCH)
> > *Cc:* barrelfish-users at lists.inf.ethz.ch
> > *Subject:* Fwd: building an arm kernel of barrelfish for gem5****
> >
> > ** **
> >
> > CCing the public barrelfish list.****
> >
> > ---------- Forwarded message ----------
> > From: *Samuel Hitz* <samuel.hitz at gmail.com>
> > Date: Mon, Feb 27, 2012 at 9:21 AM
> > Subject: building an arm kernel of barrelfish for gem5
> > To: tharris at microsoft.com
> >
> >
> > Hi Tim,****
> >
> > ** **
> >
> > I'm a CS student at ETH Zurich and for my Bachelor Thesis, I try to
> port
> > Barrelfish to a heterogeneous ARM multicore system (big.LITTLE) using
> the
> > gem5 simulator. Now the first step is to build a Barrelfish kernel
> suitable
> > for the use with gem5. I saw on the wiki, that you did this already
> for
> > x86. Now my question is, what must be modified in the build process
> to get
> > an ARM kernel for gem5?****
> >
> > I saw you're using 'molly' in order to build the kernel. How is this
> > applicable to build an ARM kernel for gem5? What do I have to
> rewrite? Or
> > is it just a matter of building an ARM Barrelfish tree and run 'make
> > m5_kernel'?****
> >
> > ** **
> >
> > Thank you for your help.****
> >
> > ** **
> >
> > Best,****
> >
> > ** **
> >
> > Samuel****
> >
> > ** **
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: https://lists.inf.ethz.ch/pipermail/barrelfish-
> users/attachments/20120227/2bcb60c2/attachment-0001.html
>
> ------------------------------
>
> _______________________________________________
> Barrelfish-users mailing list
> Barrelfish-users at lists.inf.ethz.ch
> https://lists.inf.ethz.ch/mailman/listinfo/barrelfish-users
>
>
> End of Barrelfish-users Digest, Vol 25, Issue 15
> ************************************************

-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium.  Thank you.