kinsey.moore at oarcorp.com
Tue Aug 24 20:09:02 UTC 2021
On 8/10/2021 13:39, Joel Sherrill wrote:
> On Tue, Aug 10, 2021, 9:58 AM Gedare Bloom <gedare at rtems.org
> <mailto:gedare at rtems.org>> wrote:
> On Tue, Aug 10, 2021 at 5:34 AM Kuan-Hsun Chen <c0066c at gmail.com
> <mailto:c0066c at gmail.com>> wrote:
> Hi Benson,
> Regarding Discord for RTEMS, here you go:
> https://discord.gg/TKhmGt8p <https://discord.gg/TKhmGt8p>.
> On Tue, Aug 10, 2021 at 1:21 PM Mathew Benson
> <mbenson at windhoverlabs.com <mailto:mbenson at windhoverlabs.com>>
> Is there a Slack or Discord channel to discuss RTEMS? I
> don't want to flood everybody's inbox with emails.
> As well, for this kind of development work (porting RTEMS), you
> can bring your discussions and questions over to devel at rtems.org
> <mailto:devel at rtems.org>.
> I want to port RTEMS to the Xilinx Zynq Ultrascale+ R5.
> I've taken the RTEMS training, but that was a couple years
> ago. I think I'll be fine once I can just get through the
> build system and can focus on just code, but the build
> system seems very foreign to me. There still seems to be
> either fragments of an old build system or just files that
> don't seem to serve any purpose. It would appear that
> "/spec/build/bsps/arm/xilinx-zynqmp" is where my build set
> definition begins, but then what is
> "/bsps/arm/xilinx-zynqmp/config" for?
> We haven't quite divorced ourselves from the older autotools build
> system, but new BSPs/ports (including the aarch64) are not using
> the old build system at all.
> The arm/xilinx-zynqmp is a BSP for only the 32-bit arm support of
> the Zynq Ultrascale+ MPSOC specifically tested/used on the
> Ultra96. It was contributed by Dornerworks. I guess this is a
> little confusing, because now we also have
> bsps/aarch64/xilinx-zynqmp, which provides the 64-bit support. At
> some point, we should be able to deprecate/drop the
> arm/xilinx-zynqmp with the ability to configure the
> aarch64/xilinx-zynqmp to only run in aarch32 mode.
> Probably worth starting a thread to ask Kinsey about that. I think the
> user was trying something like that at one point.
The arm/xilinx-zynqmp BSP is in a weird spot. It works, but requires
either booting from JTAG or a seldom-used configuration of FSBL. It
could also theoretically be booted from a very thin shim layer of
AArch64 code on top of a more typical AArch64 FSBL. Realistically, it
needs better documentation. Running ARMv7 RTEMS code on the A53 CPUs is
definitely possible and I have worked with someone who has tested on it
some, but still belongs in bsps/arm because it is fundamentally not
AArch64 even though it has an identical peripheral set and runs on the
same silicon. In my opinion, the only reason to run in this mode is to
reuse existing and tested binaries with only minor/trivial porting
required. If you're compiling the entire OS from scratch and
reverifying, it is much better to run AArch64 with the ILP32 ABI since
it gives you the same smaller-memory advantages as ARMv7 as well as the
advantages of AArch64 underneath the hood.
> I also think the BSP dornerworks contributed can run under Xen. So
> that would have to be accounted for
As best I can tell, that is a completely separate BSP just for running
under Xen on ARM and unrelated to any Zynq BSPs.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the users