lwIP | rtemslwip/tms570: Bringup lwIP network stack on TI TMS570LC4357 BSP (LaunchXL2-570LC43) (!48)

[Pavel Pisa (@ppisa)]

Pavel Pisa commented: https://gitlab.rtems.org/rtems/pkg/rtems-lwip/-/merge_requests/48#note_154990


Hello @ar-in0 and others, I have checked all provided commits and all changes seems to me as logical and correct. It would worth to test that even on other targets which have LwIP functional.

I have tested the build only for TMS570LS3137 now and when I find time I do more testing.

I have found problem with `TMS570_MMR_SELECT_GMII_SEL` which has been renamed to `TMS570_MMR_SELECT_MII_MODE` long time ago in the commit https://gitlab.rtems.org/rtems/rtos/rtems/-/commit/3a0dcd5ee97dd150cdea9d1d2ca06e81f58436fc and has not propagated into LwIP. But that is not @ar-in0 fault and needs to be updated anyway.

As for the pins setup, I would suggest to use symbolic pins/balls defines if they are found in [bsps/arm/tms570/include/bsp/tms570lc4357-pins.h](https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/bsps/arm/tms570/include/bsp/tms570lc4357-pins.h). But that is not as critical for now.

So generally this series represents lot of hard work and I will be happy if it reaches mainline.

As for the errors indications, it can be related to some access to initialized Flash or RAM or elsewhere and it would wort to attempt to find it. It is possible that some RAM area does not have initialized parity bits or something similar. Clear by loop or memset could help or the initial memory tests can help.

As for endianness errata on TMS570LC4357, I am curious if the problem is the same if the descriptors are placed in regular SRAM, but there would be problem with caching which are much harder to solve for descriptors then for data. I expect that EMAC's CPPI RAM is uncached. But I can be wrong. I think that we have tested use of other RAM for buffer descriptors during initial TMS570 bringup, but I am not sure about that. But generally it seems that solution is correct and in conformance with HalCoGen solution.

-- 
View it on GitLab: https://gitlab.rtems.org/rtems/pkg/rtems-lwip/-/merge_requests/48#note_154990
You're receiving this email because of your account on gitlab.rtems.org. Unsubscribe from this thread: https://gitlab.rtems.org/-/sent_notifications/4-cfj57yq0pnn9paua0sv9daw1x-m/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/20260722/5cbebe3c/attachment.htm>