STM32H743 memory map
Hi Folks,
@ALABB @Amel NASRI
The STM32H742/H743 Datasheet (DS12110 Rev 7, appears to be the latest updated one)
Page #54/357 of the Document: Table 7
(Flash memory and SRAM memory mapping for STM32H742xI/G)
--> memory map for the H743 is missing!
It describes the H742, but not the H743.
--> SRAM3 map is different!
The Reference Manual (RM0433 Rev 7, appears to be the latest updated one)
Page #135/3319
Embedded SRAM
The STM32H742xx, STM32H743/53xx and STM32H750xB devices feature:
– AHB SRAM3 is mapped at address 0x3004 0000 and accessible by all system
masters except BDMA through D2 domain AHB matrix. AHB SRAM3 can be used
as buffers to store peripheral input/output data for Ethernet and USB,.
It appears that quite some people are confused and used the table from the datasheet.
So, If you have read the Datasheet and the RM, probably one would be confused about SRAM3.
Please fix the docs at the earliest! Please do not wait for yet another human lifecycle :)
(When those changes are not needed anymore, because people would have moved on)
https://forums.freertos.org/t/freertos-tcp-for-stm32h7/7618/41
https://forums.freertos.org/t/freertos-tcp-for-stm32h7/7618/55
The datasheet is confusing, the drivers are buggy and having gross fundamental issues, people using the H7 drivers from ST are extremely frustrated and unhappy.
Initially, I thought I was the only one. A bit later, realized the issue on this forum itself. But thought that's to it. But quite a lot of your customers are really upset with ST drivers!
https://forums.freertos.org/t/freertos-tcp-for-stm32h7/7618/41
https://forums.freertos.org/t/freertos-tcp-for-stm32h7/7618/52
Before ST does more damage to itself, please have a wholehearted attempt to fix things.
Please, it's a request.
I was hoping that you folks, would accept @alister Alistair's fix for the Ethernet peripheral. But I do not know what's stopping you folks from accepting the fix. You did 2 revisions of your driver instead, without accepting such a major fix, where quite a lot of people are looking for. Alister mentioned in his post that he does not posses a board to include the Phy related code; But ST could have sent him a Nucleo board. Has ST become that stingy ? It would have helped ST more than anyone else. Previously, I've worked with some divisions of ST developing open source drivers for the Linux community. They even contributed hardware to help in the development of those drivers. It was a breeze and those drivers are now part of the mainline Linux source tree. I had to read entire threads, over a long period to understand why the issues are still open. Alister could easily have fixed the trouble for you. I guess, even now.
Please .... do something about the mess.
(BTW, The forum application is in an even more mess. The previous version before the so called fixing of the forum was easier to use and to search for threads. The new forum interface, one cannot even find anything. It's really terrible ... Just an honest opinion)
Thank You,
Manu