Importance of Bochs (and PearPC)
Erwin Rol
mailinglists at erwinrol.com
Tue Oct 19 08:43:20 UTC 2004
On Mon, 2004-10-18 at 13:45 -0500, Joel Sherrill wrote:
> I just wanted to reiterate the importance of Bochs and PearPC
> to RTEMS. Right now, each RTEMS user/developer has access to
> a limited set of test platforms. Often those platforms
> have no networking, disk storage, or graphics at all.
>
> As Ralf and I spoke about in Munich, the use of Bochs gives
> everyone a common high capability reference platform that is
> easy to use.
Maybe next time you are all in Munich I should drop by to say "hi", it
is only 30km from here (Holzkirchen).
>
> In that light, I am begging for some help here to get it
> going.
>
> + If you have knowledge of the ATA subsystem,
> please give Bochs a try for the File IO test and
> post the complete setup.
>
> + If you have networking experience with RTEMS, try
> to setup the NE2000 simulation in Bochs with
> RTEMS. Again post the complete setup.
>
I would like to that a try. I don't know to much about ATA so that will
have to wait, but networking should be doable. I still have a half port
of the FreeBSD IPX (yep there are still people that mistake IPX with
networking ;-) that i would like to finish in this way.
> PearPC has similar capabilities but RTEMS will likely need a
> special BSP for it. But it brings similar capabilities on
> a PowerPC.
Here i am still trying to figure out how to boot, yaboot isn't really
doing what i want it to do.
>
> Please, please take a stab at the above and post results.
>
I think the first stab would be BSP's for both of them ? Even though
bochs would not really need a BSP, i think it is better when it does
have its own BSP. Can anybody give reason why it should or should not
have its own BSP ?
> I am looking forward to the day where the RTEMS webserver can
> be run in a virtual PC and be demonstrated serving files from
> the virtual hard disk. :)
>
> It would also be an ideal platform to work on an ISO9660 filesystem,
> port a small web client, etc, etc.
>
> Please, please. :)
>
- Erwin
More information about the users
mailing list