RTEMS | bsps/stm32h7: make errata configurable (!184)

Cedric Berger (@cedric) gitlab at rtems.org
Wed Aug 14 14:24:37 UTC 2024




Cedric Berger commented on a discussion on spec/build/bsps/arm/stm32h7/errata.yml: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/184#note_110999

 > +SPDX-License-Identifier: CC-BY-SA-4.0 OR BSD-2-Clause
 > +build-type: script
 > +copyrights:
 > +- Copyright (C) 2024 Karel Gardas
 > +do-build: null
 > +do-configure: |

A few random comments here:
- The errata looks scary "Read access with unaligned address, such as a half-word read access starting at odd address, is not supported" but I suppose that on the M7 side, for simple code+data access (no peripherals) it should probably not matter because all the access goes through the cache, and the cache will access the FMC one full cache line at a time, correct?
- I've seen very weird things in that area, like writing stuff with the M4 and getting something slightly different on the M7 for the last bytes of every cache line, stuff like that. Never got to the bottom of it besides deciding to never share the FMC memory between M4 and M7.
- We've got some data corruption when using the rtems-libbsd ethernet stm driver with data in SDRAM. I wonder if this is related in any way (I'm going to do some testing in this area today)

-- 
View it on GitLab: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/184#note_110999
You're receiving this email because of your account on gitlab.rtems.org.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/bugs/attachments/20240814/1fa0fe1a/attachment.htm>


More information about the bugs mailing list