[RTEMS-Users] [Fwd: i960 simulation problem]

[Joel Sherrill]

Pam Renton wrote:
> I am looking at options for existing systems, which use the i960. So
> thanks very much. That is important information.
>   

When we obsoleted the i960 platform, we did it because of the tool
situation and because no one complained when we asked.

It SHOULD not be terribly difficult to build the last gcc 2.95 based
i960  toolset with the newlib version that matches the RTEMS
version you are interested in.  I haven't done it personally but wouldn't
expect that it to be that difficult.

One of Ralf's big improvements in the 4.7 tree is that the infrastructure
is now setup to handle different toolset component versions for different
architectures.

If you need to give some project details, feel free to email me privately.

--joel
> -----Original Message-----
> From: rtems-users-bounces at rtems.org
> [mailto:rtems-users-bounces at rtems.org] On Behalf Of Joel Sherrill
> Sent: October 5, 2006 9:19 AM
> To: RTEMS Users Mailing List
> Subject: [RTEMS-Users] [Fwd: i960 simulation problem]
>
> I think this is the only message of substance
> which came in during the transfer and probably
> did not get propagated.
>
> You don't mention a version but the i960sim BSP
> was never completely functional.  Even when it
> worked enough to run hello world, it did not
> implement all the i960 instructions generated by
> gcc and never had any support for a clock tick.
>
> Also gcc for the i960 is reported to be broken
> for all gcc versions >= 3.0.  That is ultimately
> why it was dropped from gcc.
>
> Combine that with running on cygwin and who really
> knows how broken everything is.
>
> Why again do you want to be on an i960?
>
>
>
>
> _______________________________________________
> rtems-users mailing list
> rtems-users at rtems.com
> http://rtems.rtems.org/mailman/listinfo/rtems-users
>