RTEMS | bsps/riscv: Update the NOEL BSPs (!1536)

[Sebastian Huber (@sebhub)]

Sebastian Huber created a merge request: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1536

Project:Branches: sebhub/rtems:up/noel-bsp to rtems/rtos/rtems:main
Author: Sebastian Huber

## Summary

Issue #5617: Noel BSPs fail to build and use source directly from another BSP

This merge request replaces !1471. It holds the seven commits of !1471,
reworked on top of the shared CLINT, PLIC and clock drivers, and two
commits of embedded brains.

It depends on !1511 and !1520. Its first seven commits are the commits of
these two merge requests. The last nine commits are the subject of this
merge request.

The noel BSPs find their devices through AMBA Plug & Play. They can use
the driver manager and the GRLIB drivers. The family has these BSPs:

- `noelgp32`, `noelgp64`, `noelmcl32` and `noelmcl64` for the standard
  NOEL-V configurations of the GRLIB IP Core User Manual

- `gr765` for the GR765 SoC

- `gr801` for the GR801 SoC

These are the changes against !1471:

- The BSPs use the shared drivers of !1520 through a noel
  `<bsp/irqimpl.h>`. The copies of `irq.c` and `clockdrv.c` are gone, so
  no noel build item names a source of the riscv BSP. The noel build
  still installs the `<tm27.h>` of the riscv BSP.

- The BSP option `RISCV_PLIC_PROBE_PRIORITY_BITS` probes the implemented
  bits of the PLIC priority registers. The GRPLIC implements fewer than
  32 bits.

- The `gr765` BSP starts without AMBA Plug & Play. BSP options give the
  CLINT, the PLIC, the GPTIMER, the APBUART interrupt, the hart count and
  the bus frequency.

- `<bsp.h>` includes no driver header, `<amba.h>` includes the Plug &
  Play headers only where a variant needs them, and the build installs
  `<grlib/io.h>`.

- `bsp_reset()` passes the semihosting arguments through register
  constraints in one asm statement.

- Each source of the BSPs is in a Doxygen group.

## Test

| Check | Result |
| --- | --- |
| Build of the six noel BSPs, uniprocessor and SMP, with `RTEMS_DEBUG` | no warning |
| Validation test suites of `riscv/gr765` SMP on SIS in the GR765 NOEL-V mode with 8 processors, on a downstream tree | 38 test suites, no failure |

**BSP** `riscv/gr765` · **Simulator** SIS with the GR765 support of rtems-sis!13

## AI Details
<!-- Make sure you have read our statement at https://www.rtems.org/generative-ai/ -->

### Prompt used

None as a single prompt.  The work was done in an interactive Claude Code
session on the RTEMS tree.  The task was to take over !1471, to rebase it
on the shared drivers of !1520 and to prepare the GR765 BSP for the
pre-qualification.  Each change was directed and reviewed.

### AI model used

Claude Opus 5.5 (claude-opus-5-5), through Claude Code.

### How AI was used for the contribution

- ( ) Formatting
- ( ) Test creation.
- (x) Code comments.
- ( ) The entire contribution was generated using AI
- ( ) AI code completion such as Copilot in VSCode.

The rework and the commit messages of embedded brains were produced in
that session under my direction.  The commit messages of Matteo Concas
are his.

### Access

I have not used a product which claims copyright in its output, and I have
legitimate access to the one I used.

-- 
View it on GitLab: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1536
You're receiving this email because of your account on gitlab.rtems.org. Unsubscribe from this thread: https://gitlab.rtems.org/-/namespace/49/sent_notifications/6-aufv72d1q23rlh4lkd3aytnj8-1d/unsubscribe | Manage all notifications: https://gitlab.rtems.org/-/profile/notifications | Help: https://gitlab.rtems.org/help


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/bugs/attachments/20261012/bd27583c/attachment-0001.htm>